Resuscitation 95 (2015) e169–e201

Contents lists available at ScienceDirect

Resuscitation

jou rnal homepage: www.elsevier.com/locate/resuscitation

Part 7: Neonatal resuscitation

2015 International Consensus on Cardiopulmonary Resuscitation and

Emergency Cardiovascular Care Science with Treatment Recommendationsଝ,ଝଝ

∗,1 1

Jonathan Wyllie (Co-Chair) , Jeffrey M. Perlman (Co-Chair) , John Kattwinkel,

Myra H. Wyckoff, Khalid Aziz, Ruth Guinsburg, Han-Suk Kim, Helen G. Liley,

Lindsay Mildenhall, Wendy M. Simon, Edgardo Szyld, Masanori Tamura,

2

Sithembiso Velaphi, on behalf of the Neonatal Resuscitation Chapter Collaborators

a r t i c l e i n f o

Keywords:

Cardiopulmonary resuscitation

Delivery room

Newborns

Introduction right-to-left shunting across the ductus arteriosus. The systemic

organs must equally and quickly adjust to the dramatic increase in

Newborn transition blood pressure and oxygen exposure. Similarly, intrauterine ther-

mostability must be replaced by neonatal thermoregulation with

The transition from intrauterine to extrauterine life that occurs its inherent increase in oxygen consumption.

at the time of birth requires timely anatomic and physiologic Approximately 85% of babies born at term will initiate spon-

adjustments to achieve the conversion from placental gas exchange taneous respirations within 10–30 s of birth, an additional 10%

to pulmonary respiration. This transition is brought about by ini- will respond during drying and stimulation, approximately 3%

tiation of air breathing and cessation of the placental circulation. will initiate respirations after positive-pressure ventilation (PPV),

Air breathing initiates marked relaxation of pulmonary vascular 2% will be intubated to support respiratory function, and 0.1%

resistance, with considerable increase in pulmonary blood flow will require chest compressions and/or epinephrine to achieve

1–3

and increased return of now-well-oxygenated blood to the left this transition. Although the vast majority of newborn infants

atrium and left ventricle, as well as increased left ventricular out- do not require intervention to make these transitional changes,

put. Removal of the low-resistance placental circuit will increase the large number of births worldwide means that many infants

systemic vascular resistance and blood pressure and reduce require some assistance to achieve cardiorespiratory stability each

year.

Newly born infants who are breathing or crying and have good

tone immediately after birth must be dried and kept warm so

ଝ

The European Resuscitation Council requests that this document be cited as to avoid hypothermia. These actions can be provided with the

as follows: Wyllie J, Perlman JM, Kattwinkel J, Wyckoff MH, Aziz K, Guinsburg baby lying on the mother’s chest and should not require separation

R, Kim H-S, Liley HG, Mildenhall L, Simon WM, Szyld E, Tamura M, Velaphi S,

of mother and baby. This does not preclude the need for clinical

on behalf of the Neonatal Resuscitation Chapter Collaborators. Part 7: Neonatal

assessment of the baby. For the approximately 5% of newly born

resuscitation: 2015 International Consensus on Cardiopulmonary Resuscitation and

infants who do not initiate respiratory effort after stimulation by

Emergency Cardiovascular Care Science With Treatment Recommendations. Resus-

citation 2015;95:e169–e201. drying, and providing warmth to avoid hypothermia, 1 or more

ଝଝ

This article has been co-published in Circulation and reprinted in Pediatrics. of the following actions should be undertaken: providing effec- ∗

Corresponding author.

1 tive ventilation with a face mask or endotracheal intubation, and

Co-chairs and equal first co-authors.

2 administration of chest compressions with or without intravenous

The members of the Neonatal resuscitation Chapter Collaborators are listed in

the Acknowledgments section. medications or volume expansion for those with a persistent heart

http://dx.doi.org/10.1016/j.resuscitation.2015.07.045

0300-9572/© 2015 European Resuscitation Council, American Heart Association, Inc., International Liaison Committee on Resuscitation. Published by Elsevier Ireland Ltd.

All rights reserved.

e170 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Fig. 1.

rate less than 60/min or asystole, despite strategies to achieve effec- Evidence evaluation

tive ventilation (Fig. 1).

The 2 vital signs that are used to identify the need for an inter- GRADE

vention as well as to assess the response to interventions are The task force performed a detailed systematic review based on

heart rate and respirations. Progression down the algorithm should the recommendations of the Institute of Medicine of the National

4

proceed only after successful completion of each step, the most Academies and using the methodological approach proposed

critical being effective ventilation. A period of only approximately by the Grading of Recommendations, Assessment, Development

5

60 s after birth is allotted to complete each of the first 2 steps, i.e., and Evaluation (GRADE) Working Group. After identification and

determination of heart rate and institution of effective ventilation. prioritization of the questions to be addressed (using the PICO [pop-

6

Subsequent progression to the next step will depend on the heart ulation, intervention, comparator, outcomes] format), with the

rate and respiratory response. assistance of information specialists, a detailed search for relevant

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e171

•

articles was performed in each of 3 online databases (PubMed, Comparison of different saturation percentiles to use for targeting

Embase, and the Cochrane Library). supplementary oxygen delivery in uncompromised and compro-

By using detailed inclusion and exclusion criteria, articles were mised premature infants

•

screened for further evaluation. The reviewers for each question Comparison of prolonged versus conventional inspiratory times

created a reconciled risk of bias assessment for each of the included to determine if the former is more effective in establishing func-

studies, using state-of-the-art tools: Cochrane for randomized con- tional residual capacity (FRC) and increasing the heart rate

7 •

trolled trials, Quality Assessment of Diagnostic Accuracy Studies Studies to determine the optimum technique for maintaining the

8

(QUADAS)-2 for studies of diagnostic accuracy, and GRADE for temperature of very low birth weight (VLBW) infants from the

observational studies that inform both therapy and prognosis time of delivery through admission to intensive care

questions.9

GRADE is an emerging consensus process that rates quality of

One small randomized study has addressed the question of

evidence and strength of recommendations along with values and

10 prophylactic endotracheal suctioning in the depressed baby with

preferences. GRADE evidence profile tables were created to facil- 17

meconium (see NRP 865), and 1 randomized trial of sustained

itate an evaluation of the evidence in support of each of the critical 18

inflation (SI) has recently been published (see NRP 804). Addi-

and important outcomes. The quality of the evidence (or confidence

tional studies addressing these critical questions are ongoing but

in the estimate of the effect) was categorized as high (where one has

were not available for the 2015 CoSTR review.

high confidence in the estimate of effect as reported in a synthesis

To achieve the goal of identifying a series of relevant ques-

of the literature), moderate (where one has moderate confidence,

tions, the Neonatal Task Force group comprising 38 members and

but there may be differences from a further elucidated truth), low

representing 13 countries met for the first time in May 2012 in

(where one has low confidence in the estimate of the effect that may

Washington, DC. At that meeting, a series of questions were iden-

be substantially different from the true effect), or very low (where

tified, researched, culled, and eventually refined into 26 questions

it is possible that the estimate of the effect is substantially different

11 at subsequent meetings by using the GRADE approach. One addi-

from the true effect). These categorizations were based on the

tional question, related to the accurate and timely detection of heart

study methodologies and the 5 core GRADE domains of risk of bias,

rate immediately after birth, was identified in December 2014 as a

inconsistency, indirectness (i.e. the population studied was not the

major gap in knowledge and was introduced as a late-breaking PICO

same as that for which the guideline will be used), imprecision of

question. The meetings since May 2012 included 3 ILCOR group

effect estimates, and other considerations (including publication

12 meetings (in Vienna, October 2012; Melbourne, April 2013; and

bias). Randomized studies start as high quality but may be down-

Banff, April 2014) and neonatal-specific ILCOR meetings (in Den-

graded for methodological quality, whereas observational or cohort

ver, CO, May 2013; Washington, DC, December 2013; Vancouver,

studies start off as low quality and can be further downgraded or

Canada, May 2014; and Washington, DC, December 2014).

upgraded depending on methodical quality or positive outcome

The literature was researched and consensus was reached on

effect.

the following issues:

Guideline users have to determine how much they can trust that

a recommendation will produce more favorable rather than unfa-

•

Optimal assessment of heart rate (NRP 898)

vorable consequences. The strength of a recommendation reflects a

•

Delayed cord clamping in preterm infants requiring resuscitation

gradient in guidance, with a clearer expectation for adherence with

(NRP 787)

strong recommendations (identified by the words we recommend)

•

Umbilical cord milking (NRP 849)

and lesser insistence in weak recommendations (identified by the

•

Temperature maintenance in the delivery room (NRP 589)

words we suggest). In addition, the direction of effect may be in favor

•

Maintaining infant temperature during delivery room resuscita-

of or against the recommendation. GRADE points to several factors

tion (NRP 599)

that may influence the strength of a recommendation, including the

•

Warming of hypothermic newborns (NRP 858)

risk-benefit balance, quality of evidence, patient values and pre-

•

Babies born to mothers who are hypothermic or hyperthermic in

ferences, and, finally, costs and resource utilization. If confidence

labor (NRP 804)

in these values and preferences is high and variability is low, it

•

Maintaining infant temperature during delivery room

is more likely that the recommendation will be strong (and vice

resuscitation—intervention (NRP 793)

versa). Recommendations, whether strong or weak, have different

•

Continuous positive airway pressure (CPAP) and intermittent

implications for patients, healthcare professionals, or healthcare

management. positive-pressure ventilation (IPPV) (NRP 590)

•

Sustained inflations (NRP 809)

•

Outcomes for positive end-expiratory pressure (PEEP) versus no

Generation of topics

PEEP in the delivery room (NRP 897)

•

T-piece resuscitator and self-inflating bag (NRP 870)

After publication of the 2010 International Consensus on Car- •

Intubation and tracheal suctioning in nonvigorous infants born

diopulmonary Resuscitation (CPR) and Emergency Cardiovascular

through meconium-stained amniotic fluid (MSAF) versus no intu-

13–15

Care Science With Treatment Recommendations (CoSTR), it

bation for tracheal suctioning (NRP 865)

was apparent that several unclear and contentious delivery room •

Oxygen concentration for resuscitating premature newborns

resuscitation issues remained. In 2012, the Neonatal Task Force

(NRP 864)

published an article titled “Neonatal Resuscitation: In Pursuit of •

2-Thumb versus 2-finger techniques for chest compression

16

Evidence Gaps in Knowledge,” in which the major gaps in knowl-

(NRP 605)

edge were identified. The following critical randomized studies •

Chest compression ratio (NRP 895)

were proposed with the goal for completion before the ILCOR 2015 •

Oxygen delivery during CPR—neonatal (NRP 738)

International Consensus Conference on CPR and Emergency Car- •

Laryngeal mask airway (NRP 618)

diovascular Care Science With Treatment Recommendations: •

Newborn infants who receive PPV for resuscitation, and use of a

device to assess respiratory function (NRP 806)

• •

Prophylactic postdelivery endotracheal suctioning versus no suc- Use of feedback CPR devices for neonatal cardiac arrest (NRP 862)

•

tioning in a depressed baby with meconium Limited resource-induced hypothermia (NRP 734)

e172 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

•

Delivery room assessment for less than 25 weeks and prognostic Values, preferences, and task force insights

score (NRP 805) There was much discussion and heated debate about the use of

•

Apgar score of 0 for 10 min or greater (NRP 896) ECG to determine heart rate. Although the data suggest that the

•

Predicting death or disability of newborns of greater than 34 ECG provides a more accurate heart rate in the first 3 min, there

weeks based on Apgar and/or absence of breathing (NRP 860) were no available data to determine how outcomes would change

•

Resuscitation training frequency (NRP 859) by acting (or not acting) on the information. Important issues were

•

Neonatal resuscitation instructors (NRP 867) raised about inappropriate interventions being implemented based

on a falsely low heart rate by pulse oximetry or auscultation that

Neonatal algorithm might be avoided if the heart rate could be determined by ECG. It

was pointed out that pulse oximetry is still very important for the

There was considerable debate with regard to modifying the measurement of saturation values to define supplementary oxygen

algorithm. The first debate related to the necessity of a timeline. needs. Introducing ECG leads in the delivery room will take time,

Many thought that a 30-s time rule was unreasonable and not as will acquiring methods to rapidly apply electrodes. In view of

evidenced based. On the other hand, because this is a global doc- these findings of false-positive readings by conventional means, we

ument, others advocated strongly that a reminder to assess and have no data on when to advise appropriate actions for bradycardia

intervene if necessary, within 60 s after birth, should be retained to detected by the conventional measures such as pulse oximetry or

avoid critical delays in initiation of resuscitation. Thus, more than auscultation. Some transient bradycardia may be normal and be

95% of newly born infants will start breathing spontaneously or reflective of timing of cord clamping. More studies are needed.

1

in response to stimulation within approximately 30 s. If apnea

persists PPV should be initiated within 60 s. As a compromise,

Knowledge gaps

the 30-s time point has been removed. Given the importance of

•

Studies delineating differences in interventions and/or patient

hypothermia as a predictor of mortality and evidence from mul-

outcomes based on ECG versus pulse oximetry measurements

tiple studies that moderate hypothermia (temperature less than

◦ •

Studies of heart rate in VLBW infants requiring resuscitation and

36 C) can be avoided with simple intervention strategies, the new

in relationship to timing of cord clamping

algorithm contains a running line reminding providers to maintain

•

Improved technology for rapid application of ECG

thermoregulation throughout the immediate newborn period.

Initial assessment and intervention Delayed cord clamping in preterm infants requiring resuscitation

(intervention) (NRP 787)

ECG/EKG in comparison to oximetry or auscultation for the

detection of heart rate (NRP 898) In preterm infants, including those who received resuscitation

(P), does delayed cord clamping (greater than 30 s) (I), compared

In babies requiring resuscitation (P), does electrocardiography with immediate cord clamping (C), improve survival, long-term

(ECG/EKG) (I), compared with oximetry or auscultation (C), mea- developmental outcome, cardiovascular stability, occurrence of

sure heart rate faster and more accurately (O)? intraventricular hemorrhage (IVH), necrotizing enterocolitis, tem-

perature on admission to a newborn area, and hyperbilirubinemia

Introduction (O)?

Neonatal resuscitation success has classically been determined

by detecting an increase in heart rate through auscultation. Heart Introduction

rate also determines the need for changing interventions and In the past 50 years, the umbilical cords of babies born preterm

escalating care. However, recent evidence demonstrates that aus- have generally been cut soon after birth, so that the newborns

cultation of heart rate is inaccurate and pulse oximetry takes can be transferred immediately to the neonatal team. However,

several minutes to achieve a signal and also may be inaccurate dur- there is recent evidence that a delay of clamping by 30–60 s

ing the early minutes after birth. This PICO question is intended to after birth results in a smoother transition, particularly if the

review the evidence regarding how best to determine heart rate baby begins breathing before the cord is cut. In both animal and

after birth. human models, the delay is associated with increased placental

transfusion, increased cardiac output, and higher and more stable

Consensus on science neonatal blood pressure. There is controversy about how long it is

For the important outcomes of fast and accurate measurement appropriate to delay clamping if the baby is perceived to require

of heart rate in babies requiring resuscitation, we have identified resuscitation.

•

Very-low-quality evidence from 5 nonrandomized studies

Consensus on science

enrolling 213 patients showing a benefit of ECG compared with

For the critical outcome of infant death, we identified very-low-

oximetry19–23

quality (downgraded for imprecision and very high risk of bias)

•

Very-low-quality evidence from 1 nonrandomized study

evidence from 11 randomized clinical trials enrolling 591 patients

enrolling 26 patients showing a benefit of ECG compared with

showing no benefit to delayed cord clamping (odds ratio [OR], 0.6; 24

auscultation 25–35

95% confidence interval [CI], 0.26–1.36).

For the critical outcome of severe IVH, we identified very-low-

The available evidence is from nonrandomized studies, down-

quality evidence (downgraded for imprecision and very high risk

graded for indirectness and imprecision.

of bias) from 5 randomized clinical trials enrolling 265 patients

showing no benefit to delayed cord clamping (OR, 0.85; 95% CI,

Treatment recommendation 0.20–3.69).26,27,31,32

In babies requiring resuscitation, we suggest the ECG can be

For the critical outcome of periventricular hemorrhage

used to provide a rapid and accurate estimation of heart rate (weak

(PVH)/IVH, we identified very-low-quality evidence (downgraded

recommendation, very-low-quality evidence).

for imprecision and very high risk of bias) from 9 randomized

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e173

clinical trials enrolling 499 patients showing benefit of delayed cord outcome, and inconsistent thresholds for treatment with pho-

26,27,29–35

clamping (OR, 0.49; 95% CI, 0.29–0.82). totherapy across studies.

For the critical outcome of neurodevelopment, we did not iden-

•

tify any evidence. Balance of consequences favors delayed cord clamping, as desir-

For the critical outcome of cardiovascular stability as assessed able consequences probably outweigh undesirable consequences

by mean blood pressure at birth, we identified very-low-quality in most settings. The results of randomized controlled trials and

evidence (downgraded for imprecision and very high risk of bias) nonrandomized observational studies with comparison groups

from 2 randomized clinical trials enrolling 97 patients show- were generally consistent. However, small and sick infants who

ing higher blood pressure associated with delayed cord clamping received immediate resuscitation were generally excluded from

29,31

(mean difference [MD], 3.52; 95% CI, 0.6–6.45). the available randomized controlled trials, so data are very

For the critical outcome of cardiovascular stability as assessed limited for this group at highest risk for physiologic instability,

by mean blood pressure at 4 h after birth, we identified very-low- complications of prematurity, and mortality who may also realize

quality evidence (downgraded for imprecision and very high risk highest benefit or harm from the intervention.

•

of bias) from 3 randomized clinical trials enrolling 143 patients Preferences (parents’) favor delayed clamping, which has

showing increased mean blood pressure at 4 h of age after delayed received strong popular support through social media and Inter-

25,31,32

cord clamping (MD, 2.49; 95% CI, 0.74–4.24). net sites. The advantages of delayed cord clamping assume

For the critical outcome of cardiovascular stability as assessed heightened importance in resource-limited settings where spe-

by blood volume, we identified very-low-quality evidence (down- cialty care for preterm neonates may be limited. Improving initial

graded for imprecision and very high risk of bias) from 2 cardiovascular stability with maintenance of temperature and

randomized clinical trials enrolling 81 patients showing benefit of lower risk of morbidities such as necrotizing enterocolitis and

35,36

delayed cord clamping (MD, 8.25; 95% CI, 4.39–12.11). severe intracranial hemorrhage may offer significant survival

For the critical outcome of temperature, on admission we iden- advantages, even where neonatal intensive care is not available.

tified very-low-quality evidence (downgraded for imprecision and In areas where maternal anemia is prevalent, iron supplemen-

very high risk of bias) from 4 randomized clinical trials enrolling tation is limited, and a safe blood supply is often unavailable, the

208 patients showing no statistically significant benefit from reduction in need for transfusion and improved blood volume at

29,31,32,34

−

delayed cord clamping (MD, 0.1; 95% CI, 0.04 to 0.24). birth may have increased significance.

For the important outcome of need for transfusion, we identi-

fied very-low-quality evidence from 7 randomized clinical trials A major debate surrounded the issue as to whether the quality

enrolling 398 patients showing less need for transfusion after of the studies was low or very low. Overall, the group thought that

28–30,32,34–36

delayed cord clamping (OR, 0.44; 95% CI, 0.26–0.75). downgrading the evidence as suggested by the GRADE tool was

For the important outcome of necrotizing enterocolitis, we not reasonable, given that this was one of the areas with the most

identified very-low-quality evidence (downgraded for impreci- randomized trial data. However, eventually based on the GRADE

sion and very high risk of bias) from 5 randomized clinical trials criteria, it was necessary to classify most of the outcomes as very-

enrolling 241 patients showing lower incidence of necrotizing ente- low-quality evidence. It was noted that the existing studies enrolled

29,31–34

rocolitis (OR, 0.3; 95% CI, 0.19–0.8). very few extremely premature infants and very few who received

For the important outcome of hyperbilirubinemia and peak resuscitation. The group was unanimous in stressing the need for

bilirubin concentrations (mmol/L), we identified moderate-quality additional research, which parallels a Cochrane review reflecting

evidence from 6 randomized clinical trials enrolling 280 patients similar sentiments of a need for more high-quality evidence. Some

showing higher peak bilirubin value in those neonates with delayed members questioned how to reconcile with obstetric guidelines,

29–33,35 37

cord clamping (MD, 16.15; 95% CI, 6.13–26.17). which has an out clause for babies requiring resuscitation.

For the important outcome of treated hyperbilirubinemia

(need for phototherapy), we identified low-quality evidence from

Knowledge gaps

1 randomized clinical trial enrolling 143 patients showing no sta-

•

Results of ongoing large randomized controlled trials

tistically significant difference (relative risk [RR], 1.29; 95% CI,

•

1.00–1.67).35 Comparison of delayed versus immediate cord clamping among

preterm infants who receive resuscitation with PPV

•

Comparison of delayed cord clamping with cord milking

Treatment recommendation •

Outcome data of high importance, such as long-term neurode-

We suggest delayed umbilical cord clamping for preterm infants velopment

not requiring immediate resuscitation after birth (weak recom- •

Need for resuscitative intervention at delivery

mendation, very-low-quality evidence). •

Hyperbilirubinemia among high-risk populations

There is insufficient evidence to recommend an approach to cord

clamping for preterm infants who do receive resuscitation imme-

Umbilical cord milking—intervention

diately after birth, because many babies who were at high risk of

requiring resuscitation were excluded from or withdrawn from the

In very preterm infants (28 weeks or less) (P), does umbil-

studies.

ical cord milking (I), in comparison with immediate umbilical

cord clamping (C), affect death, neurodevelopmental outcome at

Values, preferences, and task force insights 2–3 years, cardiovascular stability, i.e. need for pressors, need for

Overall, the quality of evidence for the question was very fluid bolus, initial mean blood pressure, IVH (any grade, severe

low. Despite drawing evidence from randomized controlled trials, grade), temperature on admission, hematologic indices (initial

the small sample size in most trials and the associated impreci- hemoglobin, need for transfusion), hyperbilirubinemia, need for

sion limited the quality of evidence for all outcomes of interest. phototherapy, or need for exchange transfusion (O)?

Although 2 larger observational trials were considered, the quality

and size of effect were not sufficient to influence the conclu- Introduction

sions. The quality of evidence for necrotizing enterocolitis and There is some evidence that “milking” of the umbilical cord

hyperbilirubinemia was limited by inconsistent definitions of the from the placental side toward the newborn may have a similar

e174 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

effect to delayed cord clamping (i.e. increased placental transfusion, recommendation was, “We suggest that cord milking, as opposed to

improved cardiac output, and increased neonatal blood pressure). immediate cord clamping, may be performed at delivery for VLBW

If correct, this would offer a more rapid alternative to delayed clam- to improve initial mean blood pressure, hematologic indices, and

ping of the cord. IVH (Grades 1 and 2).” However, concerns were raised related to the

absence of evidence pertinent to long-term outcomes and, in partic-

Consensus on science ular, neurologic outcome. Moreover, there was serious imprecision

For the critical outcome of death, we found low-quality in the data. These factors led to the final treatment recommenda-

evidence (downgraded for very serious imprecision) from 3 ran- tion.

38–40

domized clinical trials showing that there is no difference in

death (OR, 0.76; 95% CI, 0.25–2.29). Knowledge gaps

For the critical outcome of cardiovascular stability, we found

•

Evidence regarding neurodevelopmental outcomes for cord milk-

low-quality evidence (downgraded for imprecision) from 2 ran-

38,39 ing compared with immediate cord clamping is necessary.

domized studies showing that the initial mean blood pressure

•

Comparison of delayed cord clamping with cord milking

was 5.43 mm Hg higher (range, 1.98–8.87 mm Hg) in the group

•

Multiple studies of cord milking in this population are under way

receiving umbilical cord milking.

at this time, and additional data will be available in 2020.

For the critical outcome of IVH, we found low-quality evidence

(downgraded for very serious imprecision) from 2 randomized clin-

38,40 Temperature

ical trials showing a reduction of IVH (all grades: OR, 0.37;

95% CI, 0.18–0.77) but no difference (from 1 randomized clinical

38 It has been known for more than a century that preterm babies

trial in severe IVH; OR, 0.44; 95% CI, 0.07–2.76) (low-quality evi-

who become hypothermic after birth have a higher mortality

dence, downgraded for very serious imprecision) when umbilical

41

than those who remain normothermic. The association between

cord milking was performed.

hypothermia and neonatal mortality and morbidity, including

For the critical outcome of neurologic outcome at 2–3 years,

respiratory distress syndrome, metabolic derangements, IVH, and

we did not identify any evidence to address this.

late-onset sepsis, has long been recognized, with premature infants

For the important outcome of hematologic indices, we found

being particularly vulnerable (see below). Specifically, moderate

low-quality evidence (downgraded for imprecision) from 2 ran-

◦

38,39 hypothermia (temperature less than 36 C) at birth has been rec-

domized clinical trials showing that cord milking increased the

ognized as an independent risk factor for death in premature

initial hemoglobin level (MD, 2.27 g/dL; 95% CI, 1.57–2.98 g/dL) and

infants.42,43

low-quality evidence (downgraded for imprecision) from 3 ran-

38–40 These relationships reflect the fact that the premature infant

domized clinical trials showing that cord milking decreased

is at very high risk of net heat loss because of a large surface

transfusion (OR, 0.2; 95% CI, 0.09–0.44).

area-to-volume ratio and increased evaporative fluid losses from

For the important outcome of temperature, we found low-

the skin. Strategies introduced to minimize heat loss include use

quality evidence (downgraded for very serious imprecision) from

39 of occlusive wrapping, exothermic warming mattress, warmed

1 randomized clinical trial showing that the temperature of the

humidified resuscitation gases, polyethylene caps, and increasing

milking group was not different on admission.

delivery room temperature, and have met with varying success.

For the important outcome of bilirubin indices, we found

A by-product of these interventions to prevent hypothermia is

low-quality evidence (downgraded for very serious imprecision)

◦

more-frequent hyperthermia (temperature greater than 37.5 C).

showing that the maximum bilirubin measurement (3 randomized

◦

38–40 40 Hyperthermia (temperature greater than 37.5 C) also increases the

clinical trials ) and use of phototherapy (1 study ) was not

risk for neonatal mortality and morbidity in both term and preterm

different between groups.

infants. This section will review the importance of maintaining

temperature in a goal range, interventions to minimize heat loss

Treatment recommendation

at delivery, how quickly a low temperature should be raised into a

We suggest against the routine use of cord milking for infants

normal range, the impact of maternal hyperthermia and hypother-

born at 28 weeks of gestation or less, because there is insufficient

mia on the newborn, and strategies to avoid hypothermia in the

published human evidence of benefit. Cord milking may be con-

resource-limited setting.

sidered on an individual basis or in a research setting, as it may

improve initial mean blood pressure and hematologic indices and

reduce intracranial hemorrhage. There is no evidence for improve- Temperature maintenance in the delivery room—prognosis (NRP 589)

ment in long-term outcomes (weak recommendation, low-quality

evidence). In nonasphyxiated babies at birth (P), does maintenance of

◦ ◦

All studies included in this evidence review milked 20 cm of normothermia (core temperature 36.5 C or greater and 37.5 C

umbilical cord toward the umbilicus 3 times while the infant was or less) from delivery to admission (I), compared with hypother-

◦ ◦

held at the level of the introitus or below the level of the placenta mia (less than 36 C) or hyperthermia (greater than 37.5 C) (C),

before cord clamping. change survival to hospital discharge, respiratory distress, survival

to admission, hypoglycemia, intracranial hemorrhage, or infection

Values, preferences, and task force insights rate (O)?

In making this recommendation, we place a higher value on the

unknown safety profile and less value on the simplicity/economy Consensus on science

of this intervention. For the critical outcome of mortality, there is evidence from

Much of the deliberations focused on the wording of the treat- 36 observational studies of increased risk of mortality associ-

42–77

ment recommendation. The first recommendation proposed was, ated with hypothermia at admission (low-quality evidence

“We suggest that cord milking, as opposed to immediate cord but upgraded to moderate-quality evidence due to effect size,

clamping, be performed at delivery for VLBW infants.” A second dose–effect relationship, and single direction of evidence). There

recommendation was, “We suggest that cord milking, as opposed is evidence of a dose–effect relationship on mortality, suggesting

◦ ◦

to immediate cord clamping, may be performed at delivery for an increased risk of at least 28% for each 1 below 36.5 C body tem-

42,43 42,43,48,66

VLBW but should not be regarded as a standard of care.” A third perature at admission and dose-dependent effect size.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e175

78

One small randomized clinical trial (very-low-quality evidence, universal applicability, and the evidence for intervention improv-

downgraded for indirectness and serious imprecision) showed a ing respiratory outcomes over the lack of modern evidence for

reduction in adverse events, including death, intracranial hem- intervention changing mortality.

orrhage, necrotizing enterocolitis, and oxygen dependence with The group thought that this question should change to a pro-

improved temperature management, but 3 randomized controlled gnostic one. A recurring question is whether some of the babies

79–81

trials (low-quality evidence, downgraded for indirectness stay cold because of intrinsic factors. However, there are data that

and imprecision) did not show any significant improvement hypothermia upon admission impacts mortality through at least

in mortality with significantly improved temperature control. the first 6 months. It was suggested that a low temperature may

60,61,63,82

Four observational studies (very-low-quality evidence, also be related to the quality of care and environment. Most stud-

downgraded for indirectness and imprecision) did not find any ies reviewed used an axillary temperature but some older studies

improvement in mortality with improved admission temperatures, utilized a rectal temperature. The relative benefits of one over the

but they were not sufficiently powered for this outcome. other were not assessed in this PICO. The task force felt that an axil-

For the critical outcome of IVH, 8 observational studies lary temperature should be used in the delivery room but that on

(very-low-quality evidence, downgraded for risk of bias and indi- admission it should be left to individual regional practice.

◦

rectness) show hypothermia (temperature less than 36 C) in

preterm infants is associated with an increased likelihood of devel- Knowledge gaps

48,55,66,83–87

oping IVH. Eight observational studies (low-quality, •

Further studies are required to find if improved admission tem-

downgraded for indirectness) found no association between

43,60,61,88–92 perature improves mortality and other outcomes.

hypothermia and IVH.

For the important outcome of respiratory issues, there is evi-

44,48,50,67,83,93–96 Maintaining infant temperature during delivery room

dence from 9 observational studies (low-quality

resuscitation—intervention

evidence) showing an association between hypothermia and

79

respiratory disease. One large randomized controlled trial

Among preterm neonates who are under radiant warmers in the

(low-quality evidence, downgraded for imprecision and risk

hospital delivery room (P), does increased room temperature, ther-

of bias) found a reduction in pulmonary hemorrhage associ-

mal mattress, or another intervention (I), compared with plastic

ated with improved admission temperature (OR, 0.57; 95% CI, ◦

wraps alone (C), reduce hypothermia (less than 36 C) on admission

0.35–0.94). Eight observational studies (very-low-quality evi-

to neonatal intensive care unit (NICU) (O)?

dence) have shown an improvement in respiratory outcomes after

44,49,51,63,72,84,93,95

improved admission temperature maintenance.

Introduction

Two of these have shown a decrease in respiratory support with

93,96 A variety of strategies have been suggested to maintain a

improved temperature maintenance. Two observational stud-

preterm infant’s temperature; it is unknown which of these strate-

ies (very-low-quality evidence, downgraded for indirectness and

43,60 gies is/are most effective. This PICO question was intended to

imprecision) did not show any association.

identify the strategies and techniques that might be most effective.

For the serious outcome of hypoglycemia, there were seven

observational studies (very-low-quality, downgraded for risk of

bias and indirectness) showing a significant association between Consensus on science

◦ 44,67,70,97–100

hypothermia (less than 36 C) and hypoglycemia. Thermal mattress plus plastic wrap plus radiant warmer (I) versus

Two of these studies, using historical controls, showed improved plastic wrap plus radiant warmer. For the critical outcome of

44,99 ◦

glycemic control with improved normothermia. hypothermia (temperature less than 36.0 C) at NICU admission,

For the serious outcome of late sepsis, 2 observational stud- we identified low-quality evidence (downgraded for serious risk of

102

ies (very-low-quality evidence, downgraded for risk of bias and bias) from 1 randomized controlled trial enrolling 72 preterm

indirectness) indicated an association between hypothermia on infants of less than 32 weeks of gestation showing no bene-

43,101

admission and late sepsis. One observational study (low- fit to addition of a thermal mattress to the use of plastic wrap

quality, downgraded for risk of bias and indirectness) found no and radiant warmer (RR, 1.89; 95% CI, 0.18–19.95). Four observa-

66

association after multivariate analysis. tional studies (low-quality evidence, downgraded for serious risk

82,103–105

For the serious outcome of survival to admission, there is of bias) including 612 patients of less than 32 weeks of

no published evidence addressing any effect of delivery room gestation showed benefit to the addition of the thermal mattress

hypothermia upon survival to admission. (OR, 0.27; 95% CI, 0.18–0.42).

For the serious outcome of admission hyperthermia, there is For the important outcome of hyperthermia (temperature

◦

no published evidence about newborn hyperthermia at admission. greater than 38.0 C) at admission, we have identified low-quality

evidence (downgraded for serious risk of bias) from the same ran-

102 82,103,105,106

domized controlled trial and 4 observational studies

Treatment recommendations

including 426 patients showing no harm from the thermal mattress

Admission temperature of newly born nonasphyxiated infants

(RR, 3.78; 95% CI, 0.86–16.60 and OR, 6.53; 95% CI, 0.80–53.30).

is a strong predictor of mortality and morbidity at all gestations. It

should be recorded as a predictor of outcomes as well as a quality ◦

Environmental temperature 26 C or greater plus plastic wrap plus

indicator (strong recommendation, moderate-quality evidence).

radiant warmer (I) versus plastic wrap plus radiant warmer (C).

We recommend that the temperature of newly born nonas-

◦ For the critical outcome of hypothermia (temperature less than

phyxiated infants be maintained between 36.5 and 37.5 C after ◦

36.0 C) at NICU admission, we identified no studies addressing this

birth through admission and stabilization (strong recommenda-

intervention alone.

tion, very-low-quality evidence).

For the important outcome of hyperthermia (temperature

◦

greater than 38.0 C) at admission, we identified low-quality evi-

Values, preferences, and task force insights dence (downgraded for serious risk of bias) from 1 observational

107

In making these statements, we place a higher value on the study including 40 patients of less than 29 weeks of gestation

strong association of inadvertent hypothermia with mortality, the showing no harm from increasing the environmental temperature

◦

apparent dose effect, the single direction of the evidence, the 26 C or greater (OR, 8.45; 95% CI, 0.37–182.58).

e176 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Heated and humidified gases plus plastic wrap plus radiant warmer (I) potential chance for harm. A strong recommendation was made

versus plastic wrap plus radiant warmer (C). For the critical outcome because of the large numbers in the studies and the consistent

◦

of hypothermia (temperature less than 36.0 C) at NICU admission, direction of effect. There was concern about 1 randomized ther-

we identified very-low-quality evidence (downgraded for serious mal mattress trial, which was stopped for safety issues because

78

risk of bias) from 1 randomized controlled trial enrolling 203 of hyperthermia. However, this is the only study that has demon-

patients of less than 32 weeks of gestation showing no benefit (RR, strated an adverse effect with small numbers, suggesting some

0.64; 95% CI, 0.31–1.35), and 1 observational study (low-quality unclear negative (possible environmental) effect. In the treatment

108

evidence) including 112 patients of less than 33 weeks of gesta- recommendation, it was suggested to add the words may include

tion showing benefit to the use of heated and humidified gases and after the word combination.

to the use of plastic wrap and the radiant warmer (OR, 0.20; 95% CI,

0.08–0.47). Knowledge gaps

For the important outcome of hyperthermia (temperature

•

◦ Although a combination of interventions (increasing environ-

greater than 38.0 C) at admission, we identified low-quality

mental temperature, warm blankets, thermal mattress, and cap)

evidence (downgraded for serious risk of bias) from the same obser-

108 linked to quality improvement initiatives are effective in reducing

vational study showing no harm (OR, not estimable). ◦

hypothermia (less than 36 C) on NICU admission among newly

born preterm infants of less than 32 weeks of gestation who

Total body and head plastic wrap plus radiant warmer (I) versus body

are under radiant warmers and plastic wrap, the contribution of

plastic wrap plus radiant warmer (C). For the critical outcome of

◦ each intervention (increasing environmental temperature, ther-

hypothermia (temperature less than 36.0 C) at NICU admission,

mal mattress, heated and humidified gases, and cap) remains to

we identified very-low-quality evidence (downgraded for serious

109 be established.

risk of bias) from 1 randomized controlled trial enrolling 100

patients of less than 29 weeks of gestation showing no benefit to

Warming of hypothermic newborns—intervention (NRP 858)

the addition of wrapping (RR, 0.60; 95% CI, 0.24–1.53).

For the important outcome of hyperthermia (temperature

◦

In newborns who are hypothermic (temperature less than

greater than 38.0 C) at admission, we identified low-quality

◦

36.0 C) on admission (P), does rapid rewarming (I), compared with

evidence (downgraded for serious risk of bias) from the same ran-

109 slow rewarming (C), change mortality rate, short and long-term

domized controlled trial showing no harm (RR, 0.33; 95% CI,

0.01–7.99). neurologic outcome, hemorrhage, episodes of apnea and hypo-

glycemia, or need for respiratory support (O)?

◦

Combination of interventions (environmental temperature 23–25 C

plus plastic wrap without drying plus cap plus thermal mattress plus Introduction

radiant warmer) versus plastic wrap plus radiant warmer (C). For the Neonates are at high risk for becoming hypothermic during

◦

critical outcome of hypothermia (temperature less than 36.0 C) at resuscitation. Some early teaching for rewarming these neonates

admission, we identified very-low-quality evidence (downgraded has suggested that slow rewarming is preferable over faster so as

93,95,96,110

for serious risk of bias) from 4 observational studies to avoid complications such as apnea and arrhythmias. This PICO

enrolling 9334 patients of less than 35 weeks of gestation showing question is intended to review the recent evidence on this issue.

benefit from using a combination of interventions (i.e. environmen-

◦

tal temperature 23–25 C plus plastic wrap without drying plus Consensus on science

111,112

cap plus thermal mattress plus radiant warmer; OR, 0.40; 95% CI, We identified 2 randomized trials and 2 observational

113,114 ◦

0.35–0.46). studies comparing rapid (greater than 0.5 C/h) versus slow

◦

For the important outcome of hyperthermia (temperature (less than 0.5 C/h) rewarming strategies for hypothermic new-

◦ ◦

greater than 38.0 C) at admission, we have identified low-quality borns (less than 36.0 C) on admission. All studies were dated

evidence (downgraded for serious risk of bias) from 3 observational (the most recent study was published 28 years ago) and con-

93,95,110

studies enrolling 8985 patients showing no harm to the ducted in different settings (2 in low-resource countries and 2 in

combination of interventions (OR, 1.12; 95% CI, 0.82–1.52). high-resource countries); enrolled patients had different baseline

characteristics (postnatal age, gestational age, proportion of out-

Treatment recommendations born/inborn, degree of hypothermia). The quality of the studies was

Among newly born preterm infants of less than 32 weeks of very poor in terms of number of enrolled patients, inclusion criteria,

gestation under radiant warmers in the hospital delivery room, randomization methods, study design, and outcome measures.

we suggest using a combination of interventions, which may For the critical outcome of mortality, we identified low-quality

◦

include environmental temperature 23–25 C, warm blankets, plas- evidence (downgraded for serious risk of bias) from 1 randomized

112

tic wrapping without drying, cap, and thermal mattress to reduce clinical trial including 30 patients showing no benefit (RR, 0.88;

◦ 113,114

hypothermia (temperature less than 36.0 C) on admission to NICU 95% CI, 0.36–2.10) and 2 observational studies including 99

(weak recommendation, very-low-quality evidence). patients showing benefit in favor of a rapid rewarming strategy (OR,

◦

We suggest that hyperthermia (greater than 38.0 C) be avoided 0.23; 95% CI, 0.06–0.83).

due to the potential associated risks (weak recommendation, very- For the critical outcome of convulsions/seizures, we identified

low-quality evidence). very-low-quality evidence (downgraded for serious risk of bias)

112

from 1 randomized clinical trial including 30 patients show-

Values, preferences, and task force insights ing no benefit to rapid versus slow rewarming (RR, 0.88; 95% CI,

We place value on the large numbers enrolled in the observa- 0.14–5.42).

tional studies and consistent direction of effect. For the critical outcome of hemorrhage/pulmonary hemor-

Because many of the studies used multiple strategies, it was rhage, we identified very-low-quality evidence (downgraded for

112

not possible to identify the different specific interventions, that are serious risk of bias) from 1 randomized clinical trial including 30

113

effective in maintaining temperature. There was concern whether patients and 1 observational study including 38 patients show-

the recommendation should be so strong when the CIs for hyper- ing no benefit to rapid versus slow rewarming (RR, 1.31; 95% CI,

thermia (0.80–53.30) comprising 3 studies are so wide, raising the 0.26–6.76 and OR, 0.16; 95% CI, 0.02–1.50, respectively).

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e177

For the important outcome of need for respiratory support, (downgraded for risk of bias) showing an increased risk with mater-

115–121

we identified very-low-quality evidence (downgraded for serious nal hyperthermia.

114

risk of bias) from 1 observational study including 56 patients For the important outcome of adverse neurologic states

showing benefit in a slower over a rapid rewarming strategy (OR, (encephalopathy), we identified low-quality evidence from 4 non-

7.50; 95% CI, 2.14–26.24). randomized clinical trials (downgraded for risk of bias) showing an

122–125

For the important outcome of episodes of hypoglycemia, we increased risk with maternal hyperthermia.

identified very-low-quality evidence (downgraded for serious risk

of bias and very serious imprecision) from 1 randomized controlled Maternal hypothermia. For the critical outcome of mortality and

111 114

trial including 36 patients and 1 observational study includ- the important outcomes of seizures or adverse neurologic states

ing 56 patients showing no benefit to rapid versus slow rewarming (encephalopathy), we identified very-low-quality evidence from

(RR, 0.11; 95% CI, 0.01–1.81 and OR, 0.21; 95% CI, 0.01–4.06, respec- 5 randomized clinical trials (downgraded for very serious indi-

tively). rectness) that showed no significant risk of these outcomes with

126–130

For the important outcome of episodes of apnea, we identified maternal hypothermia. However, the above studies did not

very-low-quality evidence (downgraded for serious risk of bias and specifically examine these outcomes.

111,112

very serious imprecision) from 2 randomized clinical trials There are no studies of neonatal outcomes after interventions

including 66 patients showing no benefit to rapid versus slow to keep mothers normothermic.

rewarming (RR, 0.44; 95% CI, 0.04–4.32).

Treatment recommendations

Treatment recommendation Although maternal hyperthermia is associated with adverse

The confidence in effect estimates is so low that a recommen- neonatal outcomes, there is insufficient evidence to make a recom-

◦

dation for either rapid (0.5 C/h or greater) or slow rewarming mendation regarding the management of maternal hyperthermia.

◦ ◦

(0.5 C/h or less) of unintentionally hypothermic newborns (T less There is insufficient evidence to make a treatment recommen-

◦

than 36 C) at hospital admission would be speculative. dation about maternal hypothermia.

Values, preferences, and task force insights

Values, preferences, and task force insights

There was discussion as to whether this is a prognostic versus a

It was considered important to distinguish the warming of

therapeutic question. The worksheet authors used observational

infants where hypothermia is iatrogenic after birth, which in gen-

studies, because the randomized clinical trials did not focus on

eral is of a short duration, from hypothermia that is therapeutic and

the outcomes targeted. There was discussion as to whether it was

has been intentionally induced over 72 h. The latter rewarming is

possible to separate hypothermia from the cause of hypothermia.

generally recommended to be slow.

Knowledge gaps

Knowledge gaps

•

• There are no randomized controlled trials of neonatal outcomes

Attempts should be made to study a more homogenous patient

after interventions to keep mothers normothermic.

population with specific inclusion criteria stratified by gesta- •

Do interventions to achieve normothermia in mothers who are

tional and postnatal age, severity of hypothermia on admission,

hyperthermic decrease risk of adverse outcomes for newborns?

and common outcome measures.

• (Lack of randomized clinical trials)

Addressing these factors with attention to power of the study •

Do interventions to achieve normothermia in mothers who are

by using a multicenter study design will generate useful data on

hypothermic decrease risk of adverse outcomes for newborns?

which to base decisions on the rewarming strategy for hypother-

(Lack of critical/important outcomes)

mic newborns.

Maintaining infant temperature during delivery room

Babies born to mothers who are hypothermic or hyperthermic in

resuscitation—intervention (NRP 793)

labor—prognosis (NRP 804)

In newborn infants (greater than 30 weeks of gestation) in low-

In newborn babies (P), does maternal hypothermia or hyper-

resource settings during and/or after resuscitation/stabilization (P),

thermia in labor (I), versus normal maternal temperature (C),

does drying and skin-to-skin contact or covering with plastic (I),

result in adverse neonatal effects (O)? Outcomes include mortality,

compared with drying and no skin-to-skin or use of radiant warmer

neonatal seizures, and adverse neurologic states.

or incubator (C), change body temperature (O)?

Introduction Introduction

There is substantial literature from observational studies indi-

The ability to maintain temperature in a resource-limited set-

cating an association between maternal hyperthermia and neonatal

ting after birth is a significant problem (see NRP 589), with a

mortality and morbidity (see NRP 589). However, the mechanisms

dose-dependent increase in mortality for temperatures below

◦

linking these associations remain unclear. In addition, the impact of

36.5 C. Moreover, premature infants demonstrated a 12-fold

maternal hypothermia on neonatal outcome remains unclear. This

increase in mortality compared with term babies. Therefore, avoid-

PICO question attempts to address this issue.

ing hypothermia at birth would seem to be a relatively simple

intervention to reduce mortality.

Consensus on science

Maternal hyperthermia. For the critical outcome of mortality, we Consensus on science

identified low-quality evidence from 2 nonrandomized clinical tri- Plastic wraps with or without skin drying and swaddling compared

als (downgraded for risk of bias) showing an increased risk with with cot or crib with or without initial use of radiant warmer. For the

115,116

maternal hyperthermia. important outcome of normothermia or preventing hypothermia

For the important outcome of neonatal seizures, we identi- during resuscitation, we could not find any studies reporting on use

fied low-quality evidence from 7 nonrandomized clinical trials of plastic bags. During transition (from birth to 1–2 h after delivery),

e178 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

•

we identified very-low-quality evidence (downgraded for risk of Using plastic with or without drying during resuscitation

bias, inconsistency, and imprecision) from 3 randomized clinical

131–133

trials enrolling 409 newborns of greater than 30 weeks of Ventilation

gestation, showing either a reduction in incidence of hypothermia

131,132

with plastic after drying (RR, 0.77; 95% CI, 0.65–0.90) or no The respiratory management of the newly born infant in part

133

difference in temperature with plastic with or without drying depends on whether the infant is making some respiratory effort

compared with cot bed or open crib and swaddling with or without or not. In the breathing term or preterm infant, application of CPAP

initial use of radiant warmer. may be sufficient to augment endogenous effort. In the absence of

respiratory effort, establishment of FRC may be more difficult to

Skin-to-skin contact versus cot or crib with or without use of radiant establish in some cases. In the term infant, positive inflating pres-

warmer. sure may be sufficient to establish FRC, whereas in other cases PEEP

and/or an SI may be helpful. In this section, we will review the evi-

•

During transition (birth to 1–2 h after delivery), we identified dence for the use of CPAP in the spontaneously breathing infant, and

very-low-quality evidence (downgraded for risk of bias, indirect- the use of SI and/or PEEP in the nonbreathing infant. This section

134–140

ness, and imprecision) from 7 randomized clinical trials will also examine the important question of whether a nonbreath-

enrolling 600 newborns of greater than 30 weeks of gestation ing infant delivered in the presence of MSAF needs to be intubated

showing a reduction in the number of babies with hypothermia for suctioning or not. Finally, the starting oxygen concentration in

134,136,137,140

when nursed with skin-to-skin contact after delivery a premature newborn will be reviewed.

135,138,139

or similar body temperatures when compared with cot

or crib and swaddling with or without initial use of radiant CPAP and IPPV—intervention (NRP 590)

warmer.

In spontaneously breathing preterm infants with respiratory

Skin-to-skin contact versus incubator. For the important outcome of distress requiring respiratory support in the delivery room (P), does

normothermia or preventing hypothermia during resuscitation, the use of CPAP (I), compared with intubation and IPPV (C), improve

we could not find any studies reporting on skin-to-skin contact. outcome (O)?

During transition (birth to 1–2 h after delivery), we identified very-

low-quality evidence (downgraded for risk of bias, indirectness, Introduction

136,141

and imprecision) from 2 randomized clinical trials enrolling CPAP was introduced to neonatology in the 1970s for treatment

66 newborns of greater than 30 weeks of gestation showing reduc- of respiratory distress syndrome. However, because of equipment

141

tion in incidence of hypothermia by about 90% or a 50% reduction limitations, this treatment modality was not part of the early rec-

136

in drop in body temperature with skin-to-skin contact compared ommendations for neonatal resuscitation at birth. Over the past

with incubator. decade, the use of CPAP rather than the immediate intubation and

ventilation for preterm babies who do not breathe well sponta-

Treatment recommendations neously after birth has been explored. Initially, this controversy

There are no data examining the use of plastic wrap during was also complicated by the common teaching that babies born

resuscitation/stabilization. To maintain body temperature or pre- very preterm (less than 32 weeks of gestation) should be intubated

vent hypothermia during transition (birth to 1–2 h of life), we electively at birth for the purpose of administering surfactant. There

suggest that after a well newborn infant of greater than 30 weeks was also a concern that the use of CPAP in the delivery room might

of gestation has been dried, his or her trunk and limbs may be lead to a higher incidence of pneumothorax. Several randomized

put in a clean food-grade plastic bag and swaddled compared with controlled studies have tested these concerns, which prompted the

open crib or cot and swaddling (weak recommendation, very-low- following 2 PICO analyses.

quality evidence).

Consensus on science

There are no data on skin-to-skin contact during resusci-

For the critical outcome of death or bronchopulmonary dys-

tation/stabilization. To maintain normal body temperature or

plasia, we identified moderate-quality evidence (downgraded for

prevent hypothermia during transition (birth to 1–2 h after deliv-

142–144

risk of bias) from 3 randomized clinical trials enrolling 2358

ery), we suggest well newborns of greater than 30 weeks of

preterm infants born at less than 30 weeks of gestation showing

gestation be nursed with skin-to-skin contact or kangaroo mother

potential benefit to starting treatment with CPAP in the first 15 min

care compared with a cot/open crib and swaddling or incubator

after birth (RR, 0.91; 95% CI, 0.83–1.00).

(weak recommendation, very-low-quality evidence).

For the critical outcome of death, we identified moderate-

quality evidence (downgraded for risk of bias, imprecision) from

Values, preferences, and task force insights

142–144

the same 3 randomized clinical trials showing no benefit to

In making this suggestion on plastic wrap, we considered the

starting treatment with CPAP (RR, 0.82; 95% CI, 0.66–1.03). How-

decrease in hypothermia with plastic. However, clean plastic may

ever, we recognize that while the point estimate would suggest

not be available and could be costly, and use of unclean plastic may

potential for benefit, the confidence intervals cross unity to 1.03,

lead to infections.

suggesting that the potential for harm is minimal.

In making this suggestion on skin-to-skin contact, we valued the

For the critical outcome of bronchopulmonary dysplasia, we

prevention of hypothermia by using a free and effective interven-

tion. identified moderate-quality evidence (downgraded for indirect-

142–144

ness) from the same 3 randomized clinical trials showing

An issue was raised about the quality and the safety of occlu-

no benefit to starting treatment with CPAP (RR, 0.92; 95% CI,

sive wrap, and the suggestion was made to include food-grade

0.82–1.03). However, we recognize that while the point estimate

quality. The question was raised with regard to the availability of

thermometers. would suggest potential for benefit, the confidence intervals cross

unity to 1.03, suggesting that the potential for harm is minimal.

For the critical outcome of air leak, we identified very-low-

Knowledge gaps

quality evidence (downgraded for inconsistency and very serious

•

The feasibility of skin to skin during resuscitation 142–144

imprecision) from the same 3 randomized clinical trials

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e179

showing no benefit to starting treatment with CPAP (RR, 1.24; 95% The following 3 PICO analyses reflect an in-depth analysis of the

CI, 0.91–1.69). different strategies that have been suggested for this initial estab-

For the critical outcome of severe IVH, we identified very- lishment of FRC after birth.

low-quality evidence (downgraded for inconsistency and serious

142–144

imprecision) from the same 3 randomized clinical trials

Sustained inflations—intervention (NRP 809)

showing no benefit to starting treatment with CPAP (RR, 1.09; 95%

CI, 0.86–1.39).

In term and preterm newborn infants who do not establish spon-

For the important outcome of necrotizing enterocolitis, we

taneous respiration at birth (P), does administration of 1 or more

identified moderate-quality evidence (downgraded for impreci-

142–144 pressure-limited sustained lung inflations (I), compared with inter-

sion) from the same 3 randomized clinical trials showing

mittent PPV with short inspiratory times (C), change Apgar score

no benefit to starting treatment with CPAP (RR, 1.19; 95% CI,

at 5 min, establishment of FRC, requirement for mechanical venti-

0.92–1.55).

lation in first 72 h, time to heart rate greater than 100/min, rate of

For the important outcome of severe retinopathy of pre-

tracheal intubation, overall mortality (O)?

maturity, we identified low-quality evidence (downgraded for

143,144

very serious imprecision) from 2 randomized clinical trials

enrolling 1359 infants showing no benefit to starting treatment Consensus on science

with CPAP (RR, 1.03; 95% CI, 0.77–1.39). For the critical outcome of need for mechanical ventilation in

the first 72 h after birth, low-quality evidence (downgraded for

Treatment recommendation inconsistency, indirectness, and imprecision) from 3 randomized

For spontaneously breathing preterm infants with respiratory clinical trials enrolling 404 newborns showed significant benefit

18,147,148

distress requiring respiratory support in the delivery room, we of sustained lung inflations. In addition, very-low-quality

suggest initial use of CPAP rather than intubation and IPPV (weak evidence (downgraded for variability of interventions in SI and

recommendation, moderate-quality evidence). control populations) from 2 cohort studies with a total of 331

patients also showed benefit of sustained lung inflations as com-

18,149

pared with intermittent PPV with short inspiratory times. One

Values, preferences, and task force insights

151

randomized clinical trial was excluded from analysis due to

In making this suggestion, we recognize that the absolute reduc-

methodological concerns pertaining to differences in the various

tion in risk of adverse outcome associated with starting with CPAP

interventions between the study groups of which sustained lung

is small and that infants recruited to the trials had a high rate of

inflation was merely one.

treatment with antenatal steroids but we value the less invasive

approach. For the critical outcome of mortality, low-quality evidence

13–15 (downgraded for indirectness and imprecision) from 3 randomized

CPAP was introduced in the 2010 CoSTR as an option to be

18,147,149

clinical trials enrolling 404 newborns and very-low-

considered for babies who are breathing, but breathing with diffi-

quality evidence (downgraded for variability of interventions in

culty. The previous recommendation had been to simply administer

sustained lung inflation and control populations) from 2 cohort

blow-by oxygen. The current PICO question did not address the

studies with a total of 331 patients showed no benefit as compared

option of using no support. There was a consensus that, in the

18,147,149

with IPPV with short inspiratory times.

absence of contrary evidence, administration of CPAP, with or with-

For the critical outcome of bronchopulmonary dysplasia,

out supplementary targeted oxygen, is preferable in this situation

low-quality evidence (downgraded for inconsistency, indirectness,

if resources permit.

and imprecision) from 3 randomized clinical trials enrolling 404

18,147,149

patients showed no benefit. Very-low-quality evidence

Knowledge gaps

(downgraded for variability of interventions in SI and control popu-

•

The balance of risks and benefits of this approach in infants who lations) from 2 cohort studies with a total of 331 patients showed

have not received antenatal steroids is unknown. significant benefit of sustained lung inflations as compared with

• 18,149

A further trial of CPAP versus intubation and IPPV in high-risk IPPV with short inspiratory times.

preterm infants at lower gestations is required to determine the For the critical outcome of air leak, low-quality evidence (down-

risks and benefits more clearly. It is not clear whether there is a graded for inconsistency, indirectness, and imprecision) from 3

18,147,148

significant effect on mortality. The CIs for the other morbidities of randomized clinical trials enrolling 404 newborns and

prematurity leave open the possibility that any benefit in relation very-low-quality evidence (downgraded for variability of interven-

to bronchopulmonary dysplasia might still be balanced by a small tions in SI and control populations) from 2 cohort studies with a

increase in risk of severe IVH or necrotizing enterocolitis. total of 331 patients showed no effect of sustained lung inflation as

• 147,148

The utility of using an intubation-surfactant-extubation sequence compared with IPPV with short inspiratory times.

145

(INSURE) approach to facilitate early stabilization on CPAP For the important outcome of Apgar score, there was no differ-

18,147–149

soon after birth has been compared with CPAP alone in at least 2 ence between groups in any studies reviewed.

trials. This should be the subject of a future worksheet. For the important outcome of need for intubation, very-low-

quality evidence (downgraded for lack of controls) from 1 cohort

18

Ventilation strategies in the delivery room study showed that the need in the delivery room was significantly

lower in infants who received an SI compared with conventional

The most effective method for establishing an FRC in the fluid- management.

filled lung of a newborn who does not breathe spontaneously has For the important outcome of heart rate greater than 100/min,

146

been debated for many decades. In the 1980s, Vyas et al. sug- no evidence was found.

gested a technique of administering an SI of up to 5 s in duration. For the important outcome of establishment of FRC, no evi-

Both standard IPPV with or without PEEP and inflation breaths up to dence was found.

3 s in duration are currently initial strategies advocated to initiate For the important outcome of Fio2 in the delivery room, no

ventilation (Neonatal Resuscitation Program, European Resuscita- evidence was found.

tion Council). Several recent animal studies have suggested that a For the important outcome of chest compressions in the deliv-

longer SI may be beneficial for short-term respiratory outcomes. ery room, no evidence was found.

e180 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Additional comments: individual councils may choose to interpret the recommendations differently.

•

No human studies evaluated time to heart rate greater than

io

100/min, establishment of FRC, F 2 in the delivery room, or need Knowledge gaps

for chest compressions in the delivery room.

•

• The duration of an SI, the appropriate peak initial inflation pres-

In a small case series of 9 asphyxiated term infants (very-low-

sure, the number of SIs to be administered, and an early measure

quality evidence), a prolonged initial inflation of 5 s produced a

146 of response remain unclear.

2-fold increase in FRC compared with historic controls.

•

• Further studies are essential to determine the optimal pressure

Comparison of all studies (randomized clinical trials and cohort)

and duration of SI that would allow the establishment of FRC

was compromised due to the heterogeneity of methodology, i.e.

while minimizing the risk of barotrauma in the newly born infant

wide differences in duration of the initial SI (5–20 s) as well as

and long-term morbidity.

the peak inspiratory pressure (20–30 cm H2O) and use of a vari-

ety of interface devices to deliver the SI (endotracheal tube, face

Outcomes for PEEP versus no PEEP in the delivery

mask, or nasopharyngeal tube). Three studies repeated the ini-

18,149,150 room—intervention (NRP 897)

tial sustained lung inflation once, 1 at a higher positive

18

inflating pressure, whereas 1 study repeated the SI twice with

148 In preterm/term newborn infants who do not establish respira-

increasing positive inflating pressure.

• tion at birth (P), does the use of PEEP as part of the initial ventilation

No studies compared the efficacy of a single SI with multiple SIs.

• strategy (I), compared with no PEEP (C), improve Apgar score at

Animal studies of the effects of SI on alveolar recruitment have

151 152 5 min, intubation in the delivery room, chest compressions in the

shown in lambs and preterm rabbits more uniform lung

delivery room, heart rate greater than 100/min by 2 min of life,

inflation and better lung compliance, if animals received an SI

time for heart rate to rise above 100/min, air leaks, oxygen satura-

before initiation of mechanical ventilation. However, a study

153 tion/oxygenation, Fio in the delivery room, mechanical ventilation

by Klopping-Ketelaars showed no benefit after an initial SI 2

in the first 72 h, bronchopulmonary dysplasia, survival to discharge

in preterm lambs, and another study showed that stepwise

(O)?

increases in PEEP resulted in better overall lung mechanics than

154

treatment with an initial SI.

Introduction

Treatment recommendation In the 2010 CoSTR, new recommendations were introduced

We suggest against the routine use of initial SI (greater than regarding the use of CPAP for babies exhibiting breathing difficulty

5 s duration) for preterm infants without spontaneous respirations and for using PEEP whenever IPPV was required. But problems

immediately after birth, but an SI may be considered in individual have continued because of an inability of self-inflating bags to

clinical circumstances or research settings (weak recommendation, reliably deliver PEEP, and self-inflating bags are the most com-

low-quality evidence). mon devices used for neonatal resuscitation worldwide. This PICO

question and the one immediately following (NRP 870) were con-

Values, preferences, and task force insights structed to examine the value of using one device over another and

In making this recommendation, and in the absence of long- the need for PEEP when administering IPPV during resuscitation.

term benefits, we place a higher value on the negative aspect

involving lack of clarity as to how to administer sustained lung Consensus on science

inflations versus the positive findings of a reduced need for intuba- For the critical outcome of mortality before discharge, we iden-

tion at 72 h. tified low-quality evidence from 2 randomized trials of 596 preterm

Although the studies reviewed showed that administration of newborns showing no benefit (RR, 0.616; 95% CI, 0.274–1.382) to

an SI reduced the need for mechanical ventilation in the first 72 h providing PEEP compared with no PEEP (downgraded for serious

156,157

of life, the use of SI did not change the incidence of important imprecision and risk of bias).

long-term outcomes related to lung function, including risk of bron- For the critical outcome of chronic lung disease, we iden-

chopulmonary dysplasia or overall mortality. Studies thus far are tified moderate-quality evidence from 2 randomized trials of

likely underpowered for these outcomes. 596 preterm newborns showing no benefit (RR, 1.153; 95% CI,

There was much debate about the use of SI. The methods used 0.711–1.871) to providing PEEP as compared with no PEEP (down-

156,157

in delivering SI varied among studies. It was stressed that different graded for imprecision and risk of bias).

devices varied in their ability to generate pharyngeal pressures. For the critical outcome of need for cardiac drugs or chest

Moreover, a recent animal study suggests that there may be unin- compressions in the delivery room, we identified low-quality evi-

tended glottis closure associated with SI. There was also concern dence from 2 randomized trials of 596 preterm newborns showing

that the current wording of the treatment recommendation may be no benefit (RR, 1.468; 95% CI, 0.550–3.917) to providing PEEP as

156,157

viewed by some as limiting the potential for future clinical studies. compared with no PEEP (downgraded for imprecision and

Evidence evaluators were asked to decide whether to include risk of bias).

155

the te Pas article. The decision was made to exclude it because For the important outcome of oxygen saturation at 5 min after

of multiple confounding interventions. It was thought that more birth, we identified moderate-quality evidence from 1 randomized

detail in the consensus on science was needed to reflect that studies trial of 80 preterm newborns showing no benefit (P = 0.55) to pro-

used SI ranging from 5 to 25 s. There was debate about the use of viding PEEP (median Spo2, 49%; interquartile range [IQR], 25–90%)

the wording suggest against. Several members were in favor of using versus not providing PEEP (median Spo2, 59%; IQR, 33–66%) (down-

156

this term, because there is insufficient evidence regarding how to graded for imprecision and risk of bias).

administer sustained lung inflation, how many such breaths should For the important outcome of maximum concentration of oxy-

be applied, or whether it should be used with or without PEEP. It gen used during resuscitation, we identified low-quality evidence

is difficult to extrapolate from animal data, because the animals from 1 randomized trial of 516 preterm newborns showing mod-

in the studies were nonbreathing and had tracheostomies, so that erate benefit (P = 0.005) to providing PEEP (mean, 48%; standard

the anatomy, physics, and physiology are different. Although there deviation [SD], 0.2) versus not providing PEEP (mean, 53%; SD,

157

was consensus agreement on the current wording, it was noted that 0.2).

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e181

For the important outcome of heart rate greater than 100/min positive effect found was a 5% change in Fio2 (see comments after

at 2 min of age, we identified low-quality evidence from 1 random- NRP 870).

ized trial of 516 preterm newborns showing no benefit to providing

PEEP versus not providing PEEP (RR, 1.656; 95% CI, 0.938–2.923)

157 Knowledge gaps

(downgraded for imprecision and risk of bias).

•

For the important outcome of time for heart rate to rise to Properly powered, well-designed randomized trials specifically

greater than 100/min, we identified moderate-quality evidence addressing important outcomes for the effects of PEEP in the

from 1 randomized trial of 516 preterm newborns showing no delivery room are necessary.

•

benefit to providing PEEP (median, 1 min; IQR, 0.5–1.8) versus It remains unclear as to the optimal level of PEEP to use.

•

not providing PEEP (median, 1 min; IQR, 0.5–1.9) (downgraded for The question of static PEEP versus dynamic PEEP needs to be

157

imprecision and risk of bias). delineated.

•

For the important outcome of need for intubation in the Differential effects of PEEP at different gestational ages and for

delivery room, we identified moderate-quality evidence from 2 different pathologies remain to be determined.

randomized trials of 596 preterm newborns showing no benefit

(RR, 1.208; 95% CI, 0.907–1.609) to providing PEEP (downgraded

156,157 T-piece resuscitator and self-inflating bag—intervention (NRP 870)

for imprecision and risk of bias) versus not providing PEEP.

For the important outcome of need for mechanical ventila-

In newborns (preterm and term) receiving ventilation (PPV)

tion in the first 72 h, we identified low-quality evidence from 1

during resuscitation (P), does using a T-piece resuscitator with

randomized trial of 80 preterm newborns showing no benefit (RR,

PEEP (I), compared with using a self-inflating bag without PEEP (C),

0.317; 95% CI, 0.093–1.086) to providing PEEP (downgraded for

achieve spontaneous breathing sooner and/or reduce the incidence

imprecision and risk of bias) versus not providing PEEP. We identi-

157 of pneumothorax, bronchopulmonary dysplasia, and mortality (O)?

fied only 1 randomized clinical trial that included term infants,

which provided insufficient data to address this question as a sec-

ondary outcome measure in a subgroup analysis (very-low-quality Introduction

evidence, downgraded for serious imprecision and risk of bias). The T-piece resuscitator has replaced the self-inflating and flow-

For the important outcome of pulmonary air leaks, we identi- inflating bag in many institutions. One major reason for this change

fied low-quality evidence from 2 randomized trials of 596 preterm has been the inability of the self-inflating bag to deliver either CPAP

newborns showing no benefit (RR, 1.401; 95% CI, 0.414–4.735) to or PEEP reliably. Advantages of the T-piece include ease of use and

156,157

providing PEEP (downgraded for imprecision and risk of bias) ability to deliver CPAP, PEEP, and/or IPPV. However, it also requires

versus not providing PEEP. a pressurized-gas source to drive the device. This PICO question is

For the important outcome of Apgar score less than 6 at 5 min, intended to review the evidence of the utility of self-inflating bags

we identified moderate-quality evidence from 1 randomized trial versus T-piece resuscitators.

of 516 preterm newborns showing no benefit to providing PEEP

(RR, 0.813; 95% CI, 0.472–1.402) (downgraded for imprecision and

157 Consensus on science

risk of bias) versus not providing PEEP.

For the following consensus on science statements, the analysis

For the less-important outcome of Apgar score at 5 min, we 156

is based on all patients (n = 80) from 1 study and from a subgroup

identified moderate-quality evidence from 1 randomized trial of 157

analysis (n = 453) in a second study.

80 preterm newborns showing no benefit (P = 0.18) to providing

For the critical outcome of death before discharge, we iden-

PEEP (median, 7; IQR, 6–8) versus no PEEP (median, 7; IQR, 6–9)

156 tified low-quality evidence (downgraded for risk of bias and

(downgraded for imprecision and risk of bias). 156,157

imprecision) from 2 randomized clinical trials enrolling 532

patients showing no benefit to the use of a T-piece resuscitator as

compared with a self-inflating bag (OR, 0.68; 95% CI, 0.31–1.56).

Treatment recommendations

For the critical outcome of bronchopulmonary dysplasia,

We suggest using PEEP ventilation for premature newborns

which was only assessed for infants of less than 1500 g, we

during delivery room resuscitation (weak recommendation, low-

identified low-quality evidence (downgraded for risk of bias and

quality evidence). 156,157

imprecision) from 2 randomized clinical trials enrolling 151

We cannot make any recommendation for term infants because

patients showing no benefit to the use T-piece resuscitator as com-

of insufficient data.

pared with self-inflating bag (OR, 0.92; 95% CI, 0.59–1.43).

For the critical outcome of air leaks, we identified low-quality

evidence (downgraded for risk of bias and imprecision) from 2 ran-

Values, preferences, and task force insights 156,157

domized controlled trials enrolling 532 patients showing no

In making this suggestion, we are considering the small

benefit to the use of T-piece resuscitator as compared with self-

reduction in maximum oxygen concentration needed during resus-

inflating bag (OR, 1.72; 95% CI, 0.51–5.78).

citation with 5 cm H2O PEEP compared with those not receiving

For the important outcome of achieving spontaneous breath-

PEEP shown in 1 human study, and considering the evidence from

ing or reducing intubation in delivery room, we identified

animal studies (see NRP 809). Interpretation of human studies is

very-low-quality evidence (downgraded for risk of bias, impreci-

further complicated by varying interfaces (e.g., face mask versus 156,157

sion, and inconsistency) from 2 randomized clinical trials

endotracheal tube) and methods of generating PEEP (e.g., self-

enrolling 532 patients showing no benefit to the use of T-piece

inflating bags with PEEP valve versus T-piece resuscitator).

resuscitator as compared with self-inflating bag (OR, 0.80; 95% CI,

Only 1 study was available to indirectly address the specific PICO

157 0.59–1.07).

question, where a subgroup comparison was applied. Good ani-

mal studies are available but are classified as low levels of evidence

from the point of applicability due to indirectness (see NRP 809). Treatment recommendation

There was concern that the evidence based on the GRADE criteria There is insufficient evidence, so the recommendation of one

was regarded as low quality. There was a major struggle to come device over another would be purely speculative because the con-

up with a recommendation when the evidence was weak. The only fidence in effect estimates is so low.

e182 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Values, preferences, and task force insights and/or incidence of MAS (including depressed and/or vigorous

The current studies suggest a benefit to using PEEP to assist infants) when infants born through MSAF were intubated for

establishment of an FRC during transition of the fluid-filled lung to tracheal suctioning (downgraded for indirectness).

an air-breathing organ. However, the evidence to date is not suffi-

ciently compelling to recommend against using a self-inflating bag Treatment recommendation

(in which reliable administration of PEEP is not achievable with cur- There is insufficient published human evidence to suggest rou-

rent devices) during neonatal resuscitation, particularly in regions tine tracheal intubation for suctioning of meconium in nonvigorous

where pressurized gases are not readily available. PEEP is recom- infants born through MSAF as opposed to no tracheal intubation for

mended when the facilities and equipment permit it to be given suctioning.

reliably (approximately 5 cm H2O).

Values, preferences, and task force insights

Knowledge gaps In making this suggestion, we place value on both harm avoid-

ance (delays in providing bag-mask ventilation, potential harm of

• 157

One cluster randomized controlled trial showed benefit of

the procedure) and the unknown benefit of the intervention of

using T-piece resuscitator for achieving spontaneous breathing

routine tracheal intubation and suctioning.

in the late preterm (mean gestational age 36 weeks) population.

Routine suctioning of nonvigorous infants is more likely to result

Further research in this population would be important.

• in delays in initiating ventilation, especially where the provider

There are no studies comparing the flow-inflating bag to either

is unable to promptly intubate the infant or suction attempts are

the self-inflating bag or the T-piece resuscitator (with or without

repeated. In the absence of evidence of benefit for suctioning, the

PEEP) for neonatal resuscitation. Theoretically, the flow-inflating

emphasis should be on initiating ventilation within the first minute

bag should be similar to the T-piece resuscitator, although ease

of life in nonbreathing or ineffectively breathing infants.

of use may prove it to be less effective.

• Much of the deliberations focused on the wording of the

Studies comparing the flow-inflating bag to the other 2 devices

treatment recommendation. There were 3 different treatment

would be helpful.

recommendation options. First “We suggest against the routine

intubation of nonvigorous infants born through MSAF.” Second “We

Intubation and tracheal suctioning in nonvigorous infants born

suggest that routine tracheal intubation for suctioning of meco-

through MSAF versus no intubation for tracheal

nium in nonvigorous infants should not be considered as a standard

suctioning—intervention (NRP 865)

of care but may be considered a reasonable alternative to no tra-

cheal intubation in some settings.” Third “We suggest that routine

In nonvigorous infants at birth born through MSAF (P), does

tracheal intubation for suctioning of meconium in nonvigorous

tracheal intubation for suctioning (I), compared with no tracheal

infants should not be considered as a standard of care but may

intubation (C), reduce meconium syndrome or prevent death (O)?

be considered a reasonable alternative to no tracheal intubation if

a meconium plug is suspected.” There was concern that the legal

Introduction

profession could misinterpret the term standard of care. Consensus

For more than 30 years, it has been recommended that new-

was reached on the final treatment recommendation.

borns with MSAF should receive endotracheal intubation, with

tracheal suctioning using the endotracheal tube as a suction device.

Knowledge gaps

Approximately 15 years ago, as a result of a multicenter random-

•

Tracheal intubation or no tracheal intubation for suctioning in

ized clinical trial, the recommendation was restricted to babies who

nonvigorous infants: Is there a benefit or harm?

appeared to have respiratory compromise at birth (i.e. were non-

vigorous). It remains controversial as to whether even nonvigorous

babies benefit from this procedure. This PICO question is intended Oxygen concentration for resuscitating premature

to address this issue. newborns—intervention (NRP 864)

Among preterm newborns (less than 37 weeks of gestation)

Consensus on science

who receive PPV in the delivery room (P), does the use of high

For the critical outcome of mortality and/or meconium aspira-

O (50–100%) as the ventilation gas (I), compared with low

tion syndrome (MAS), we identified 1 randomized study involving 2

concentrations of O (21–30%) (C), decrease mortality, decrease

122 infants (low-quality evidence, downgraded for risk of bias and 2

17 bronchopulmonary dysplasia, decrease retinopathy, decrease IVH

imprecision) comparing tracheal intubation for suctioning versus

(O)?

no tracheal intubation for suctioning in nonvigorous infants show-

ing no benefit to suctioning in either reduced mortality and/or MAS.

Introduction

For the critical outcome of mortality and/or MAS, we identi-

158–160 The fact that high oxygen concentrations can be toxic to the

fied very-low-quality evidence from 3 studies including 12

newly born lungs has been recognized in all CoSTR statements

389 MSAF infants showing higher incidence of MAS in depressed

since 2000. The original studies examined only 21% oxygen versus

infants (268/1022, 26%) who had tracheal intubation for suctioning

100% and led to a recommendation that blended oxygen be used

compared with vigorous infants (34/11 367, 0.3%) who were not

to titrate the concentration to achieve an oxygen saturation that is

intubated (downgraded for indirectness).

reflective of what healthy babies born at term experience (i.e. tar-

For the critical outcome of mortality and/or MAS, we identi-

161–167 geted saturation). There has been an ongoing controversy as to what

fied evidence from 7 very-low-quality observational studies

the initial oxygen concentration should be. Babies born at term

demonstrating improved survival and lower incidence of MAS

should be started in air (21%), but there has been uncertainty as to

when infants (including depressed and/or vigorous infants) born

whether the preterm baby should be started in a high concentration

through MSAF were intubated for tracheal suctioning (downgraded

(50–100%) versus low concentration (21–30%) of oxygen while the

for indirectness and inconsistency).

pulse oximetry is being attached. This PICO question was intended

For the critical outcome of mortality and/or MAS, we

to examine only the starting concentration of administered oxygen,

identified evidence from 9 very-low-quality observational

158–160,168–173 not the targets.

studies demonstrating no improvement in survival

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e183

Consensus on science Knowledge gaps

For the critical outcome of mortality before discharge, we •

The most appropriate time-specific oxygen targets for premature

found moderate-quality evidence from 7 randomized clinical

newborns need to be defined.

trials enrolling 607 subjects showing no benefit to beginning •

Neurodevelopmental outcomes for preterm newborns resusci-

resuscitation with high-oxygen as compared with low-oxygen con-

tated with low- and high-oxygen concentrations need to be

centration (RR, 1.48; 95% CI, 0.8–2.73). The quality of evidence

174–180 determined.

was downgraded for imprecision. When limited to random-

ized clinical trials with concealed allocation and oxygen targeting

as a cointervention, we found moderate-quality evidence from Circulatory support

5 trials enrolling 468 subjects showing no benefit to beginning

resuscitation with a high-oxygen concentration as compared with Circulatory support focused on the most effective method of

low-oxygen concentration (RR, 1.33; 95% CI, 0.68–2.62). The qual- delivering chest compressions and included comparison of the 2-

175,177–180

ity of evidence was downgraded for imprecision. We thumb versus the 2-finger techniques as well as comparing various

found very-low-quality evidence from 1 cohort study including 125 compression-to-ventilation ratios. During the evidence evalua-

subjects showing no benefit to beginning resuscitation with high- tion in 2010, it was decided to continue recommending a chest

oxygen as compared with low-oxygen concentration (RR, 1.31; 95% compression-to-ventilation ratio of 3:1 as opposed to 15:2 or 30:2,

CI, 0.41–4.24). The quality of evidence was downgraded for serious predominantly because profound bradycardia or asystole in the

181

imprecision. newly born period is invariably secondary to an asphyxial rather

For the critical outcome of bronchopulmonary dysplasia, we than a primary cardiac event. Evidence in this review was sought

found low-quality evidence from 5 randomized trials enrolling 502 to determine whether there was any recent evidence to change this

subjects showing no benefit to beginning resuscitation with a high- recommendation. Moreover, factors important to the ergonomics

oxygen as compared with low-oxygen concentration (RR, 1.08; 95% of CPR for enhancing blood flow during chest compressions were

CI, 0.59–1.98). The quality of evidence was downgraded for incon- identified. The evidence below summarizes these findings.

175,177–180

sistency and imprecision.

For the critical outcome of intraventricular hemorrhage,

2-Thumb versus 2-finger techniques for chest

we found moderate-quality evidence from 4 randomized clini-

compression—intervention (NRP 605)

cal trials enrolling 400 subjects showing no benefit to beginning

resuscitation with a high-oxygen as compared with low-oxygen

In neonates receiving cardiac compressions (P), does the use

concentration (RR, 0.90; 95% CI, 0.47–1.72). The quality of evidence

175,178–180 of a 2-thumb technique (I), compared with a 2-finger technique

was downgraded for imprecision.

(C), result in return of spontaneous circulation (ROSC), improved

For the important outcome of retinopathy of prematurity,

neurologic outcomes, improved survival, improved perfusion and

we found moderate-quality evidence from 3 randomized trials

gas exchange during CPR, and decreased compressor fatigue (O)?

enrolling 359 subjects showing no benefit to beginning resuscita-

tion with a high- as compared with low-oxygen concentration (RR,

1.28; 95% CI, 0.59–2.77). The quality of evidence was downgraded Introduction

175,178,179

for imprecision. Two different techniques for administering chest compressions

during resuscitation of neonates have been suggested: 2 thumbs,

with fingers surrounding the lateral and posterior chest, versus 2

Treatment recommendations fingers placed vertically on the lower sternum. This PICO question

We recommend against initiating resuscitation of preterm is intended to evaluate which technique is preferable.

newborns (less than 35 weeks of gestation) with high supplemen-

tary oxygen concentrations (65–100%).

Consensus on science

We recommend initiating resuscitation with a low-oxygen

For the critical outcomes of time to ROSC, survival rates, or

concentration (21–30%) (strong recommendation, moderate-

neurologic injury, we found no data.

quality evidence).

For the critical outcome of improved perfusion and gas

exchange during CPR, we identified low-quality evidence from

9 randomized controlled trials (downgraded for indirectness and

Values, preferences, and task force insights 182–190

imprecision) and 6 nonrandomized controlled trials (down-

In making this recommendation, we place value on not exposing 191–196

graded for indirectness, imprecision, and high risk of bias)

preterm newborns to additional oxygen without proven benefit for

identifying higher blood pressure generation with the 2-thumb

critical or important outcomes. Our preference for each outcome,

versus the 2-finger method.

therefore, was to describe the risk of high-oxygen relative to low-

For the important outcome of compressor fatigue, we identified

oxygen concentration. In all studies, irrespective of whether air or

low-quality evidence from 4 randomized controlled trials (down-

high oxygen including 100% was used to initiate resuscitation, by 183,197

graded for indirectness and imprecision), with 2 identifying

the time of stabilization most infants were in approximately 30%

less fatigue with the 2-thumb versus the 2-finger technique, and 2

oxygen. We recognize that all but 1 included study allowed adjust- 189,198

studies finding no difference.

ment of oxygen concentration based on pulse oximetry and/or

New compression methods:

heart rate response.

Concerns were expressed about the practical implications of

• 199

recommending separate and simultaneous monitoring of both Thumb and index finger (TIF) compared the new method

heart rate and oxygen saturation, although accurate measure- versus the 2-thumb and 2-finger methods on manikins. Cardiac

ments of both variables are important (see NRP 898). The compressions lasted for only 5 min while recording rate, hand

chosen range for the low oxygen starting point (21–30%) location, depth, incomplete recoil, excessive depth, and error

was also questioned, but the available articles defined it. rate during CPR. Two-thumb and TIF had less decay in “suitable

Whether the high oxygen should be greater than 60% was also chest compressions” over the 5 min compared with the 2-finger discussed. method.

e184 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

• 200

Adhesive glove compared using the adhesive glove with Summary: The lower one third of the sternum remains the best

conventional CPR in 4 groups, including an infant group in a location to press over the newborn heart. Superimposed thumbs

manikin model. The 2-thumb method was used as standard in may be the better technique.

the infant group versus adhesive 2-thumb method. The theory is

that the glove enables active compression-decompression. Rate, Treatment recommendations

compression, and decompression depth were measured. No dif- We suggest that chest compressions in the newborn should

ferences in fatigue variables were found amongst groups. Results be delivered by the 2-thumb, hands-encircling-the-chest method

showed more active decompression with the adhesive glove as the preferred option (weak recommendation, very-low-quality

group. evidence).

We suggest that chest compressions should be delivered over

the lower third of the sternum (weak recommendation, very-low-

Summary: No evidence was found supporting the new thumb

quality evidence).

and index finger technique as superior to the 2-thumb method. The

adhesive glove enhanced active decompression but did not reduce

fatigue. Values, preferences, and task force insights

None are noted.

Other issues:

Knowledge gaps

• 201

Does the CPR technique cause fractures? Franke performed a •

No studies of any kind regarding the most critical outcomes were

10-year retrospective survey to determine whether the 2-thumb

available.

technique causes rib fractures. All infants received CPR plus chest •

No data from good transitional models were found.

X-rays. Median age was 9 days. •

There are very limited human neonatal data.

Summary: There was no evidence of rib fractures in any case.

Chest compression ratio—intervention (NRP 895)

•

Best location on the sternum: Using 4 assessment methods In neonates receiving cardiac compressions (P), do other ratios

202

over a wide age range of infants, it was confirmed that the (5:1, 9:3, 15:2, synchronous, etc.) (I), compared with 3:1 com-

heart lies under the lower third of the sternum. In addition, pressions to ventilations (C), increase survival rates, improve

blood pressure readings were higher when cardiac compres- neurologic outcomes, improve perfusion and gas exchange dur-

sions were applied to the lower versus the middle third of the ing CPR, decrease time to ROSC, decrease tissue injury, or decrease

sternum. Use of the infant computed tomography (CT) scan compressor fatigue (O)?

data (mean age, 4.4 months) and adult thumb side-by-side

203

measurements on manikins confirmed that the left ventri- Introduction

cle lies mostly under the lower quarter of the sternum. No Chest compressions administered in a ratio of 3 compressions to

functional data were collected to confirm better outcomes if 1 ventilation have been recommended for resuscitation of neonates

compressions focused on that area. An assumption was made at birth. The concept has been that newborns are born with

that the lower third of the sternum was the best position for lungs filled with fluid, much of which is absorbed directly across

204

compressions. the alveolar membrane with the first few breaths. If a newborn

•

Term and preterm babies: Correct positioning on the chest was is compromised sufficiently to prevent spontaneous breathing,

determined to be much better with the 2-thumb method in both resulting in bradycardia or cardiac arrest, successful resuscitation

groups of babies, although incorrect placements were found for must achieve adequate lung aeration and ventilation to reverse an

both techniques in infants less than 1500 g. Chest X-ray analy- asphyxial pathophysiology. Thus, the focus of newborn resuscita-

205

sis of term and preterm babies found the heart to be under tion efforts must be primarily aimed at establishing ventilation first

the lower third of the sternum. Chest CT scans of infants (mean and cardiac support second. This PICO question is meant to iden-

age, 4.7 months), compared with adult thumb measurements tify which compression-to-ventilation ratio will be most effective

on a manikin, comparing the 2-thumb method side by side or at achieving this.

206

superimposed, demonstrated that the side-by-side method

increases the likelihood of other organs (lungs and liver) being Consensus on science

under the points of compressions application. A manikin study Animal studies demonstrate no advantage to higher

looked at fatigue levels with the 2-thumb technique, comparing compression-to-ventilation ratios (very-low-quality evidence,

207

side-by-side or superimposed thumb position demonstrated downgraded for potential bias, indirectness, and imprecision)

that the superimposed thumb technique generated higher sim- regarding

ulated blood pressure and pulse pressure but had a higher

•

fatigue-rating score. Physiologic indices of fatigue showed no Short-term survival (2 randomized controlled trials including 54

210,211

difference between groups. CT scans of the chest to compare pigs)

•

thumb (side-by-side)/fingers measurements placed on manikins Gas exchange during CPR (2 randomized controlled trials includ-

210,211

were conducted to determine which method avoided com- ing 54 pigs)

•

pressing other structures when using the lower third of the Time to ROSC (2 randomized controlled trials including 54

208 210,211

sternum. Both methods compress other structures, but the pigs)

•

2-thumb method (side-by-side) performs better than the two Markers of tissue injury (lung/brain) (2 randomized controlled

212,213

finger method. The accuracy of using the nipple line to the trials including 54 pigs)

xiphisternum landmarks for 2-finger chest compression was

209

examined by Clements. They concluded that this method

There was no evidence identified to address the critical issue of

could result in abdomen and xiphisternum compression in

neurologic outcome.

all infants and suggested an alternate method of determining

Manikin studies demonstrated a disadvantage to

position.

higher compression-to-ventilation ratios (5:1, 9:3, 15:2)

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e185

(very-low-quality evidence, downgraded for potential bias, as guided by pulse oximetry. However, once the resuscitation has

imprecision, and indirectness) with regard to reached the need for chest compressions, it has been suggested

to increase the Fio2. This PICO question is intended to consider

•

Compressor fatigue (better depth of compression, less decay evidence to determine if this is the correct or incorrect practice.

in depth over time; 1 randomized controlled trial including 32

214

resuscitation providers) Consensus on science

•

Minute ventilation (1 randomized controlled trial including 32 For the critical outcome of ROSC, we found 8 animal studies

214 216–223

resuscitation providers) (lambs/pigs/rats) all demonstrating no advantage to 100%

•

A single manikin study demonstrated higher minute ventilation over 21% during CPR (very-low-quality evidence, downgraded for

for asynchronous compressions (120 compressions: 40 ventila- bias and indirectness).

tions) compared with 3:1 (90 compressions: 30 ventilations) (1 For the critical outcome of survival, we found 8 of 9 animal

randomized controlled trial including 2 resuscitation providers studies (lambs/pigs/rats) reporting on survival demonstrated no

215 216–223

with 5 different sessions per treatment arm) advantage to 100% over 21% during CPR. However, 1 study

(mouse) of 9 studies evaluating this outcome found an advantage to

224

Treatment recommendation 100% O2 (very-low-quality evidence, downgraded for potential

We suggest continued use of a 3:1 compression-to-ventilation bias, inconsistency, and indirectness). All studies combined showed

ratio for neonatal CPR (weak recommendation, very-low-quality 80/100 (80%) versus 74/102 (73%) survival for 100% O2 versus air

evidence). (not different). Eight studies with no advantage showed 70/77 (91%)

versus 71/79 (90%) survival. One study with advantage for 100%

showed 10/23 (43%) versus 3/23 (13%) survival (P = 0.02).

Values, preferences, and task force insights

For the critical outcome of neurologic outcome, we found 4

We prefer to retain our prior recommendation of 3:1

218,221,222,224

animal studies (pigs/rats/mice) reporting on neuro-

compression-to-ventilation ratio for neonatal CPR, because there

logic outcome with varying results (very-low-quality evidence,

is no compelling evidence suggesting a benefit to other ratios for

downgraded for potential bias, inconsistency, indirectness, and

the newborn. Since asphyxia is the predominant cause of cardio-

imprecision). One demonstrated no difference in neurologic deficits

vascular collapse in the newborn, effective resuscitation requires

218

at 72 h, and ischemic neurons in hippocampal were not different.

significant focus on ventilation. In addition, we value consistency

One demonstrated worse 4-h neurologic examination in the 100%

in the resuscitation algorithm and education programs unless new

221

O group. One demonstrated more hippocampal apoptosis in

evidence drives the change. 2

222

the 100% O group. One demonstrated more rapid restoration

All studies were done in young posttransitioned piglets (no 2

of cerebral blood flow but no difference in histologic brain injury

human or animal data in a transitioning model). Since there is no

scores.224

evidence in either a human or animal with fluid-filled lungs, we

For the critical outcome of oxidative injury, we found

need to be clear when communicating with other groups (pedi-

10 animal studies reported on oxidative injury with varying

atrics and basic life support providers) that neonates have unique

212,213,216,219–223,225–227

results (very-low-quality evidence, down-

cardiopulmonary physiology, prompting our unique 3:1 ratio.

graded for potential bias, inconsistency, and indirectness).

Some may not agree, but the values and preferences statement

Six studies (pigs/mice) demonstrated no difference in various

expresses why we still favor a 3:1 ratio.

212,213,219–221,224

oxidative injury markers, 3 (lambs/rats) demon-

strated more oxidative damage from using 100% O including Knowledge gaps 2

216,222,226

apoptosis, and a pig study reported less striatal and hip-

•

Specific research is required, such as clinical and appropriate ani- 227

pocampal apoptosis with 100% O2 compared with 21% O2.

mal model studies.

•

We need neonatal human data. Treatment recommendation

•

How many compressions in a row are required to achieve forward

There are no human data to inform this question.

blood flow and adequate coronary perfusion pressure during

Despite animal evidence showing no advantage to the use of

newborn asphyxial arrest?

100% oxygen, by the time resuscitation of a newborn baby has

•

How many interposed ventilations are needed to achieve and

reached the stage of chest compressions, the steps of trying to

maintain normocapnia during cardiac compressions due to new-

achieve ROSC using effective ventilation with low-concentration

born asphyxial arrest?

oxygen should have been attempted. Thus, it would seem prudent

•

Asynchronous technique deserves more investigation.

to try increasing the supplementary oxygen concentration (Good

•

Is ventilation adequate with SI cardiac compressions? Practice Guidance).

•

How should we limit interruptions in compressions to assess effi-

If used, supplementary oxygen should be weaned as soon as the

cacy?

heart rate has recovered (weak recommendation, very-low-quality

evidence).

Oxygen delivery during CPR (neonatal)—intervention (NRP 738)

Values, preferences, and task force insights

In neonates receiving cardiac compressions (P), does 100% O2 Although most of the available animal evidence suggests that

as the ventilation gas (I), compared with lower concentrations of resuscitation using air during neonatal chest compressions is feasi-

oxygen (C), increase survival rates, improve neurologic outcomes, ble and that 100% O2 as the resuscitation gas may increase oxidative

decrease time to ROSC, or decrease oxidative injury (O)? injury, we remain concerned that we have no human data to prove

feasibility and none of the animal studies have evaluated use of

Introduction room-air CPR for more than brief asystole. We value balancing

Neonatal resuscitation has historically focused on achieving the desire to prevent ongoing hypoxic injury in these profoundly

adequate oxygenation as quickly as possible. Recently, it has been asphyxiated neonates with the desire to prevent subsequent hyper-

recognized that excessive oxygen administration can be toxic. oxic injury.

Current guidelines recommend starting resuscitation with low This was a much-debated topic. In the case of hypotension and

inspired oxygen and then increasing inspired oxygen as necessary bradycardia, the experimental evidence is clear: You only need to

e186 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

use room air. Thus, in this case, we are making the recommendation identified low-quality evidence (downgraded for very serious risk

228–230

independent of the evidence. Perhaps, we say, “Despite no evidence, of bias) from the same randomized clinical trials showing

for the following reasons, we recommend. . ..” In training scenarios, that the laryngeal mask was more effective than the face mask

once chest compressions are started, failing to turn up O2 is a com- (OR, 0.13; 95% CI, 0.05–0.34).

•

mon error of the learner. But is it a serious error? The indirectness For the critical outcome of increasing Apgar score, we have iden-

does not inform the recommendation. We are not even following tified low-quality evidence from the same randomized controlled

low-level animal evidence. We are making a conscious decision to trials (downgraded for very serious risk of bias); the method of

take no notice of the evidence. Can we say why this group values reporting precluded analysis of this outcome.

•

giving oxygen for asystole? The task force considered the option We did not identify any evidence to address the critical outcomes

of making a neutral recommendation (with either 21% or 100% O2) of indicators of brain injury or long-term outcomes.

•

and allowing councils to decide what to do. Is this a place where For the important outcome of morbidity (gastric distention or

we do not want to suggest air or oxygen? We have no data, but we vomiting), we identified low-quality evidence (downgraded for

need to say something. imprecision and very serious risk of bias) from the same random-

228–230

ized clinical trials showing no difference for any variable

Knowledge gaps between the laryngeal mask and the face mask (OR, 5.76; 95% CI,

• 0.7–47.32).

Specific research is required, i.e. studies in good transitional ani-

mal model of asphyxia-induced severe bradycardia or asystole

For comparison of laryngeal mask to endotracheal tube as a sec-

and any neonatal human data.

ondary device (i.e. laryngeal mask or intubation when bag-mask

ventilation has failed) for infants at term requiring PPV for resusci-

Assist ventilation devices and CPR feedback devices

tation, we identified the following evidence (1 randomized clinical

231

trial with 40 patients) :

There are numerous techniques used and advocated to ventilate

effectively. In addition there are devices used to assess respiratory •

For the critical outcome of achieving vital signs or successful

function and to provide feedback during CPR. The following reviews

resuscitation, we identified very-low-quality evidence (down-

were undertaken to assess the role of alternative techniques to ven-

graded for imprecision, risk of bias) from 1 randomized clinical

tilate effectively when intubation is not feasible or unsuccessful and 231

trial showing that laryngeal mask airway was as effective as

to ascertain the evidence of feedback devices on resuscitation skill

the endotracheal tube.

performance and outcomes. •

For the critical outcome of need for subsequent endotracheal

intubation after failed bag-mask ventilation, we identified very-

Laryngeal mask airway—intervention (NRP 618)

low-quality evidence (downgraded for imprecision, risk of bias)

231

from the same randomized clinical trial showing that the

In newborn infants at near term (greater than 34 weeks) or term laryngeal mask was as effective as the endotracheal tube.

•

who have indications for intermittent positive pressure for resus- For the critical outcome of increasing Apgar score, we identified

citation (P), does use of a laryngeal mask as a primary or secondary very-low-quality evidence (downgraded for imprecision and risk

231

device (I), compared with mask ventilation or endotracheal intu- of bias) from the same randomized clinical trial ; the method

bation (C), improve response to resuscitation or change outcome of reporting precluded analysis of this outcome.

•

(O), including indicators of neonatal brain injury, achieving stable For the critical outcome of mortality, we identified very-low-

vital signs, increasing Apgar scores, long-term outcomes, reduc- quality evidence (downgraded for imprecision and risk of bias)

231

ing the need for subsequent intubation, or neonatal morbidity and from the same randomized clinical trial showing no difference

mortality? between the laryngeal mask or the endotracheal tube.

•

We did not identify any evidence to address the critical outcome

Introduction of indicators of brain injury or long-term neurologic outcomes

Endotracheal intubation is the most difficult skill to learn and comparing laryngeal mask airway or endotracheal tube as a sec-

teach in neonatal resuscitation. The laryngeal mask has recently ondary device.

•

been suggested as an alternative, either as a primary device, replac- For the important outcome of morbidity, we identified very-low-

ing face-mask ventilation, or as a secondary device for failed quality evidence (downgraded for imprecision and risk of bias)

231

or not-possible endotracheal intubation. This PICO question is from the same randomized clinical trial showing more trauma

intended to review the evidence for the utility and efficacy of the to tissue when comparing laryngeal mask versus endotracheal

laryngeal mask for neonatal resuscitation. tube (OR, 2.43; 95% CI, 0.51–11.51).

Consensus on science Treatment recommendations

For comparison of laryngeal mask airway to face mask as a pri- We suggest the laryngeal mask may be used as an alternative

mary device (i.e. use of laryngeal mask ventilation rather than to tracheal intubation during resuscitation of the late-preterm and

bag-mask ventilation for infants at term requiring PPV for resus- term newborn (more than 34 weeks) if ventilation via the face mask

citation) we identified 3 randomized controlled trials enrolling a is unsuccessful (weak recommendation, low-quality evidence).

total of 469 patients: In the unusual situation where intubation is not feasible after

failed PPV, the laryngeal mask is recommended for resuscitation of

•

For the critical outcome of achieving vital signs, we identi- the late-preterm and term newborn (more than 34 weeks) (strong

fied low-quality evidence (downgraded for very serious risk of recommendation, good clinical practice).

bias) from 2 small randomized clinical trials and 1 large quasi-

228–230

randomized clinical trial showing that the laryngeal mask Values, preferences, and task force insights

was more effective than the face mask (OR, 11.43; 95% CI, In making these recommendations, we place a moderate value

4.01–32.58). in the proven safety and feasibility for a laryngeal mask to provide

•

For the critical outcome of need for subsequent endotra- ventilation in newborns while recognizing the necessity for more

cheal intubation after failed laryngeal mask or face mask, we studies in other clinical settings (e.g., premature infant). We also

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e187

place high value on the idea that an alternative airway is a poten- Values, preferences, and task force insights

tially lifesaving intervention when face-mask ventilation has failed We should consider revising future PICO questions to embrace

and/or endotracheal intubation is unsuccessful or not feasible. new technologies for more reasonable outcomes and benchmarks

There is now reasonable evidence to add a recommendation for rather than death and disability. It was stressed that it is important

the late-preterm infant. to point out the human factors piece of the equation. The devices are

only as useful as how well the human care provider can interface

Knowledge gaps with and incorporate them appropriately into care. Another point

• raised is that we have process outcomes, but do they impact actual

The effectiveness and safety of laryngeal mask airway com-

performance? Do we need this to be a more stepwise approach?

pared with mask ventilation as the primary interface in term and

What other process outcomes should be included? In the future,

preterm infants; insertion technique, which model, and how to

we need to look at device design, types of alarms (visual or audio,

teach its use

color, font, etc.). If this were a medication, we would suggest against

something with such resource implications.

Newborn infants who receive PPV for resuscitation, and use of a

device to assess respiratory function—diagnostic (NRP 806)

Knowledge gaps

•

In newborn infants who receive PPV for resuscitation (P), does There is a need for large studies powered for important clinical

use of a device to assess respiratory function with or without pres- outcomes to determine the role of flow and volume monitor-

sure monitoring (I), compared with no device (C), change survival ing and capnography in improving response to and outcomes of

to hospital discharge with good neurologic outcome, IVH, time newborn resuscitation.

•

to heart rate greater than 100/min, bronchopulmonary dysplasia, There is a need for further research to determine whether routine

pneumothorax (O)? use of flow and volume monitoring for task training in newborn

resuscitation improves training or clinical outcomes.

•

Introduction There is a need for specific research to determine whether con-

Resuscitation of babies at birth often involves assisting venti- tinuous monitoring of flow and volume or exhaled CO2 levels

lation with positive-pressure devices. Current guidelines for this compete with other essential auditory and visual cues that need

technique have always involved recommending a specific pressure to be appreciated and responded to by resuscitation teams.

range to inflate the lungs. Recent research has indicated that exces-

sive pressure can seriously injure the lungs, particularly in babies Use of feedback CPR devices for neonatal cardiac

born preterm, and some have advocated that resuscitation guide- arrest—diagnostic (NRP 862)

lines should be based on volume rather than pressure. It has also

been suggested that measuring exhaled CO2 might indicate ade- In asystolic/bradycardic neonates receiving cardiac compres-

quate ventilation. Devices for measuring both of these variables sions (P), does the use of feedback devices such as end-tidal carbon

ETCO )

have been developed. This PICO question is meant to assess the dioxide ( 2 monitors, pulse oximeters, or automated compres-

advisability of recommending their use during resuscitation. sion feedback devices (I), compared with clinical assessments of

compression efficacy (C), decrease hands-off time, decrease time

Consensus on science to ROSC, improve perfusion, increase survival rates, or improve

Flow and volume monitoring. For the critical outcome of survival neurologic outcomes (O)?

to hospital discharge and IVH, we identified low-quality evidence

(downgraded for risk of bias and imprecision) from 1 pilot random- Introduction

232

ized controlled trial enrolling 49 babies showing no benefit. The current measure for determining successful progress in

For the critical outcome of time to heart rate greater than neonatal resuscitation is to assess the heart rate response. Other

100/min and neurologically intact survival, we found no evidence. devices such as CO2 monitoring and pulse oximetry have been sug-

For the important outcome of bronchopulmonary dysplasia gested as more sensitive measures. This PICO question is designed

and pneumothorax, we found no evidence. to determine the current evidence regarding this issue.

Capnography. For the critical outcome of survival to hospital dis- Consensus on science

charge and IVH, we identified low-quality evidence (downgraded For the critical outcomes of improved perfusion, decreased time

for risk of bias and imprecision) from 1 pilot randomized clinical to ROSC, decreased hands-off time, increased survival rates, or

233

trial enrolling 48 babies showing no evidence. improved neurologic outcomes, we found no specific data.

For the critical outcome of time to heart rate greater than Increased exhaled CO2: Five small observational studies (2

234,235

100/min and neurologically intact survival, we found no evidence. piglet posttransitioned models), 2 dog posttransitioned

236,237

For the important outcome of bronchopulmonary dysplasia models (these latter 2 articles were the identical sample of

and pneumothorax, we identified low-quality evidence (down- dogs and data but published in separate journals), and 1 human

238

graded for risk of bias and imprecision) from 1 pilot randomized study of very low quality (downgraded for indirectness and risk

233

clinical trial enrolling 48 babies showing no evidence. of bias) assessed the ETCO2 levels associated with the onset or

presence/absence of ROSC.

Treatment recommendations

Although a feasible technique, we suggest against the routine • 234 236,237

One piglet study and the dog studies associated the pres-

use of flow and volume monitoring for babies who receive PPV at

ence of decreased time to ROSC with an ETCO2 of 27–28 mm Hg.

birth, until more evidence becomes available (weak recommenda-

CPR in these studies was started after 5–10 min of cardiac arrest.

tion, low-quality evidence). • 235

One piglet study associated the presence of a heart rate greater

Although a feasible technique, we suggest against the routine

than 60/min with an ETCO2 of 14 mm Hg (sensitivity, 93%; speci-

use of capnography for babies who receive PPV at birth, until more

ficity, 81%). CPR was started at onset of asystole.

evidence becomes available (weak recommendation, low-quality •

One human study covered a wide age range of children, 1 week

evidence). 238

to 10 years. The majority were out-of-hospital arrests. ETCO2

e188 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

levels in all patients who did not attain ROSC never rose above Consensus on science

15 mm Hg. For the critical outcome of death or disability, we identified

very-low-quality evidence (downgraded for risk of bias and indi-

239,240

rectness) from 2 randomized controlled trials enrolling 338

Treatment recommendation

infants showing benefit to the use of therapeutic hypothermia (OR,

In asystolic/bradycardic neonates, we suggest against the rou-

0.43; 95% CI, 0.26–0.7).

tine reliance on any single feedback device such as ETCO2 monitors

For the critical outcome of death to latest follow-up, we iden-

or pulse oximeters for detection of ROSC until more evidence

tified very-low-quality evidence (downgraded for risk of bias,

becomes available (weak recommendation, very-low-quality evi-

dence). inconsistency, and indirectness) from 4 randomized controlled

239–242

trials enrolling 416 infants showing no benefit to the use

of therapeutic hypothermia (OR, 0.72; 95% CI, 0.44–1.16).

Values, preferences, and task force insights

Several questions were raised: Should detection of ROSC be the Treatment recommendations

only real outcome for the question because identifying this is the We suggest that newly born infants at term or near-term with

first step to recovery? Thus, it is a critical tool for determining if evolving moderate-to-severe hypoxic–ischemic encephalopathy in

your actions are effective or if you need to consider other interven- low-income countries and/or other settings with limited resources

tions. Was there a need to rate the effectiveness of the equipment as may be treated with therapeutic hypothermia (weak recommen-

the critical outcome, or is the effect on the patient what is impor- dation, low-quality evidence).

tant? Does the device measure what it is supposed to measure? Cooling should only be considered, initiated, and conducted

What about human factors issues? Can providers effectively use under clearly defined protocols with treatment in neonatal care

the equipment? Does it impact outcome? facilities with the capabilities for multidisciplinary care and

availability of adequate resources to offer intravenous therapy,

respiratory support, pulse oximetry, antibiotics, anticonvulsants,

Knowledge gaps

and pathology testing. Treatment should be consistent with the

•

There is a need for large studies powered for important clinical protocols used in the randomized clinical trials in developed

outcomes to determine the role of flow and volume monitor- countries, i.e. cooling to commence within 6 h, strict temperature

◦

ing and capnography in improving response to and outcomes of control at 33–34 C for 72 h and rewarming over at least 4 h.

newborn resuscitation.

•

There is a need for further research to determine whether routine Values, preferences, and task force insights

use of flow and volume monitoring for task training in newborn In making this recommendation, we place a higher value on the

resuscitation improves training or clinical outcomes. demonstrated effectiveness of simple cooling methods and the lack

•

There is a need for specific research to determine whether con- of harm associated with these methods over the paucity of evidence

tinuous monitoring of flow and volume or exhaled CO2 levels specific to resource-limited settings.

compete with other essential auditory and visual cues that need It is difficult to define a low-resource setting. Even within a

to be appreciated and responded to by resuscitation teams. country (e.g., India) resources may vary widely. Simple methods

of cooling are successful in lowering body temperature. There was

a concern that passive cooling may not be so harmless (e.g., extreme

Postresuscitation management

hypothermia, inappropriate hypothermia). Low-resource areas do

not have nursing care to monitor the babies closely.

ILCOR previously reviewed postresuscitation strategies that

focused on glucose control and the implementation of therapeu-

Knowledge gaps

tic hypothermia to minimize or avoid reperfusion injury from

•

intrapartum hypoxia–ischemia in well-resourced settings. For this Further adequately powered randomized controlled trials of sim-

cycle, we only reviewed the potential role of therapeutic hypother- ple methods of cooling in resource-limited settings are required

mia to minimize or avoid reperfusion injury from intrapartum to improve the quality of evidence relating to this question.

•

hypoxia–ischemia where resources are limited. Specific regional guidelines should take account of public health

system priorities for allocation of available resources and the

availability of sufficient nursing and ancillary resources to safely

Limited-resource-induced hypothermia—intervention (NRP 734)

and effectively deliver cooling therapy in the facility.

In term infants with moderate/severe hypoxic–ischemic ence-

Discontinuing resuscitation

phalopathy managed in resource-limited countries (P), does

therapeutic hypothermia to core temperature of approximately

◦ Deciding how long resuscitative efforts should continue in a

33.5 C for 72 h delivered by passive hypothermia and/or ice packs

newly born infant with no heart rate and/or absent respirations

(I), versus standard therapy (C), improve the rates of death, neu-

with a very low heart rate after sustained resuscitative efforts

rodevelopmental impairments at 18 months to 2 years (O)?

remains a critically important and difficult management decision.

In recent years, long-term outcomes have shown some improve-

Introduction ment.

Therapeutic hypothermia has been shown to reduce mortality

and morbidity in term and near-term newborns who have had Delivery room assessment for less than 25 weeks and prognostic

a hypoxic–ischemic insult and are at risk for evolving encepha- score (NRP 805)

lopathy. This therapy has generally been restricted to developed

countries where resources and regional systems permit the ther- In extremely preterm infants (less than 25 weeks) (P), does

apy to be administered under a strict protocol. This PICO question delivery room assessment with a prognostic score (I), compared

is intended to determine if therapeutic hypothermia can practically with gestational age assessment alone (C), change survival to 18–22

and effectively be practiced in countries with limited resources. months (O)?

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e189

Introduction Knowledge gaps

Antenatal assignment of prognosis for survival and/or disability •

Insufficient or absent data concerning timing of death, i.e. early

of the neonate born extremely preterm has generally been made

versus later death

on the basis of gestational age alone. Recently, scoring systems for •

Lack of information on factors other than gestational age known

including additional variables such as gender, use of maternal ante-

before birth

natal steroids, and multiplicity have been developed in an effort to •

Limited information on use of combined antenatal and postnatal

improve prognostic accuracy. This PICO question was developed to

information

examine the utility of these systems. •

Inability to fully distinguish between outcomes driven by practice

(e.g., belief that mortality is universal below a certain gestational

age), surrogate decision making by parents, and physiologic lim-

Consensus on science

itations

There is no evidence that addresses the clinical prospective use

of prognostic scoring (the use of composite survival data using ges-

Apgar score of 0 for 10 min or longer—prognosis (NRP 896)

tational age and other parameters) in infants of less than 25 weeks

of estimated gestational age.

In infants with a gestational age of 36 weeks or greater and an

There is increasing retrospective evidence that prognostic accu-

Apgar score of 0 for 10 min or longer, despite ongoing resuscitation

racy is improved by using additional information such as birth

(P), what is the rate of survival to NICU admission and death or

weight, appropriateness of weight for gestational age, use of mater-

243–247 neurocognitive impairment at 18–22 months (O)?

nal antenatal steroids, multiplicity, and gender (low-quality

evidence), but there are no prospective studies showing the post-

natal effect of such improved accuracy in predicting outcome. Introduction

There has been an ongoing controversy as to how long after one

has been attempting resuscitation after birth, and a heart rate can-

Treatment recommendation not be detected, should one continue or discontinue resuscitation

There is insufficient evidence to support the prospective use of efforts. The balance must be between ceasing too early, when ROSC

any delivery room prognostic score presently described over esti- and long-term survival may still be achievable, and continuing too

mated gestational age assessment alone in preterm infants of less long, when ROSC may occur, but early death or an unacceptable

than 25 weeks of gestation. No score has been shown to improve degree of neurologic injury may have occurred. The Apgar score of 0

the ability to estimate the likelihood of survival through either 30 has classically been the criterion, because it indicates no detectable

days or in the first 18–22 months after birth. signs of life. The recommended duration of resuscitative efforts

In individual cases, when constructing a prognosis for survival after birth has variously been 15, and more recently 10 min, after

at gestations below 25 weeks, it is reasonable to consider vari- birth.

ables including perceived accuracy of gestational age assignment, The controversy has been generated from the following uncer-

the presence or absence of chorioamnionitis, and the level of care tainties: (1) It is often not clear whether resuscitation efforts have

available for location of delivery. It is also recognized that decisions taken place throughout the 10-min period, (2) There may be ques-

about appropriateness of resuscitation below 25 weeks of gesta- tions about whether the score has indeed been 0 throughout the

tion will be influenced by region-specific guidelines established by 10 min and not just at 10 min, and (3) Have resuscitation efforts

regional resuscitation councils. been optimal throughout the 10 min? Recently, the 10-min guide-

line has been subjected to further controversy, with published

reports from therapeutic hypothermia trials of an increasing num-

Values, preferences, and task force insights ber of intact survivors after 10 min of an Apgar score of 0.

In making this statement, we put a higher value on the lack of

evidence for a generalizable prospective approach changing impor-

Consensus on science

tant outcomes over improved retrospective accuracy and locally

For the critical outcome of death up to 22 months, very-low-

validated counseling policies. For antenatal counseling, the most

quality evidence (downgraded for risk of bias, inconsistency, and

useful data would give outcome figures for babies alive at the onset

imprecision) from 6 studies encompassing 8 case series showed

of labor, not just those born alive or admitted to the neonatal inten-

that 75 of 129 infants (58%) with an estimated gestational age of

sive care unit. In reality, many are already using such extended data

36 weeks or greater and an Apgar score of 0 at 10 min of life died

in antenatal counseling to try to provide parents and healthcare 248–253

before 22 months of age. Results from 3 of these studies per-

professionals with the most accurate estimates for mortality (and

formed after 2009 that included nested observational series of cases

morbidity).

from 3 randomized clinical trials of therapeutic hypothermia and

It would obviously be preferable if there were studies to show

a series of infants who received therapeutic hypothermia outside

that using such data can prospectively improve the outcome for

a randomized trial (low-quality evidence, downgraded for risk of

these babies: Does using the most accurate information have a

bias) found that 46 of 90 infants (51%) with an Apgar score of 0 at

positive influence on the difficult decisions made about whether 250,251,253

10 min died before 22 months of age.

intensive care should be implemented?

For the critical outcome of death or moderate/severe neurode-

There was agreement to amend the treatment recommendation

velopmental impairment at 22 months of age or older, 6 studies

to include consideration of possible inaccuracy of gestational age

(very-low-quality evidence, downgraded for risk of bias, incon-

assessment, as well as to include evaluation for chorioamnionitis,

sistency, indirectness, and imprecision) showed this outcome in

and level of subsequent care that may be available. A question was

106 of 129 infants (85%) with a gestational age of 36 weeks or

raised with regard to the fact that we included weights in previ- 248–253

greater and an Apgar score of 0 at 10 min of life. Results

ous statements about prognosis; however, those were taken out

from 3 of these studies performed after 2009 (very-low-quality evi-

to allow councils to make independent recommendations. Should

dence, downgraded for risk of bias, inconsistency, indirectness, and

antenatal steroids be mentioned in the treatment recommenda-

imprecision) that included nested observational series in random-

tion? The list may become exhaustive as more factors are added

ized clinical trials of therapeutic hypothermia and series of infants

(e.g., gender).

who received therapeutic hypothermia showed that this adverse

e190 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

•

outcome occurred in 68 of 90 infants (76%) with an Apgar score of Number of infants born in the study or the transferring centers

0 at 10 min. Among the 44 survivors of these studies, 22 (50%) sur- with asystole at 10 min in whom delivery room resuscitation was

vived without major/moderate disabilities. Among the 56 cooled attempted and unsuccessful

•

infants in these studies, 15 (27%) survived without major/moderate Data regarding the quality and extension of resuscitation of these

250,251,253

disabilities (very-low-quality evidence, downgraded for infants

•

risk of bias). Only a prospective international registry with all needed infor-

No studies differentiated between severe and moderate disabil- mation of infants with asystole/heart rate less than 60/min after

ity. 10 min of adequate resuscitation may provide evidence of suffi-

None of the studies described the resuscitation procedures that cient scientific merit to answer this prognostic question

were provided.

Predicting death or disability of newborns of greater than 34 weeks

based on Apgar and/or absence of breathing—prognosis (NRP 860)

Treatment recommendation

An Apgar score of 0 at 10 min is a strong predictor of mortal-

In newborn infants of greater than 34 weeks of gestation, receiv-

ity and morbidity in late-preterm and term infants. We suggest

ing PPV at birth in settings where resources are limited (P), does

that, in babies with an Apgar score of 0 after 10 min of resuscita-

presence of heart rate with no spontaneous breathing or Apgar

tion, if the heart rate remains undetectable, it may be reasonable

scores of 1–3 at greater than 5 min predict mortality or morbidity

to stop resuscitation; however, the decision to continue or discon-

or cerebral palsy (O)?

tinue resuscitative efforts should be individualized. Variables to be

considered may include whether the resuscitation was considered

Introduction

to be optimal, availability of advanced neonatal care, such as ther-

The Apgar score is intended to be a retrospective predictor of

apeutic hypothermia, specific circumstances before delivery (e.g.,

outcome, particularly at 5 min of age. It has been suggested that an

known timing of the insult), and wishes expressed by the family

Apgar score of 0 at 10 min of age is an indication to consider discon-

(weak recommendation, very-low-quality evidence).

tinuing resuscitation efforts (see NRP 896), but there have been no

other levels of Apgar assessment by which one might make discon-

tinuation decisions, such as Apgar score of 3 or less at 20 min. This

Values, preferences, and task force insights

PICO question is intended to review the recent evidence regarding

In making this statement in infants of 35 weeks or greater with

these additional predictors.

an Apgar score of 0 for 10 min or longer, the likelihood of dying

or having severe or moderate developmental disabilities at 18–24

months is very high. Studies that included 69 infants with an Apgar Consensus on science

score of 0 at 10 min after birth who were successfully resuscitated Apgar score at 20 min. For all the outcomes, we could not find stud-

and randomly assigned to hypothermia or normothermia, and case ies that reported on individual Apgar scores (1, 2, or 3) beyond

series of 21 additional infants who were managed with therapeu- 10 min. One very-low-quality study (downgraded for indirectness)

tic hypothermia, suggest improvement in outcome compared with reported on Apgar scores at 20 min but included patients with an

254

previously reported cohorts. Among these 90 infants, 45 (50%) died Apgar score of 0. This study reported that in babies weighing

and 22 (24%) survived without major or moderate disability at greater than 2500 g with an Apgar score of 0–3 at 20 min, the mor-

18–24 months. However, the number of infants with no heart rate tality was 59%, and 57% of survivors developed cerebral palsy.

at 10 min who died in the delivery room is unknown.

This topic resulted in a long and spirited debate. A question was

Apgar score at 10 min. For the critical outcome of death, we iden-

raised as to how can we say that we should consider stopping with

tified low-quality evidence (downgraded for imprecision) from 2

a 24% possibility of survival without major handicap? Is 10 min suf-

randomized studies involving babies who participated in induced-

ficient time to make this decision? It was suggested not to use the 251,255 251

hypothermia studies. One study reported mortality of

word adequate, because the resuscitation was not assessed. What

64%, 47%, and 39% for Apgar score of 1, 2, and 3, respectively, with an

would the adults do with 20% chance of survival? However, it was 255

OR of 1.42 (95% CI, 1.19–1.69) at 18–22 months. The other study

pointed out that it is not a 20% chance, because not all babies got

reported outcomes from the same study, but at 6–7 years. Babies

to cooling. Someone advocated using the term discontinue instead

with Apgar scores of 1, 2, and 3 had mortality rates of 67%, 43%, and

of withdraw. The term adequate caused a lot of debate. What do we

27%, respectively, if they were managed with induced hypothermia

mean by it? Can it be clearer? Concern was expressed that providers

and 63%, 57%, and 62% if they were not cooled.

will likely not use science to guide the decisions for this situation

For the critical outcome of moderate/severe disability, we iden-

and will likely use their own judgment. Parents tend to choose con-

tified low-quality evidence (downgraded for imprecision) from 2

tinuation even when the data are presented to them. The decision

randomized studies involving babies who participated in induced

to continue or discontinue should be based on consultation with the 251,255 251

hypothermia studies, one reporting the outcome in 50%,

family. The optimal way to restore circulation can be in the quali-

63%, and 38% for Apgar scores of 1, 2, and 3, respectively, with an

fier. An Apgar score of 0 at 10 min is a strong predictor of disability 255

OR of 1.30 (95% CI, 1.06–1.58) at 18–22 months. The other study

at all gestations.

reported at 6–7 years of life that 100%, 75%, and 9% of babies with

Apgar score of 1, 2, and 3, respectively, had moderate/severe dis-

ability if managed with induced hypothermia and 67%, 67%, and

Knowledge gaps

71% if not managed with hypothermia, although the sample size

The major flaw in the available scientific evidence regarding

was small.

outcome of term neonates with asystole after 10 min of adequate

resuscitation is the absence of data regarding

No spontaneous respiration. For the critical outcome of death, we

identified very-low-quality evidence (downgraded for impreci-

•

Number of infants born in the study centers or the transferring 256,257

sion) from 2 observational studies that time to spontaneous

centers with asystole at 10 min who were not actively resusci-

respiration of more than 30 min was associated with 52–77% mor-

tated (in the hypothermia studies many were transfers) tality.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e191

For the critical outcome of cerebral palsy or abnormal national steering committee of the US Neonatal Resuscitation Pro-

neurologic findings, we identified very-low-quality evidence gram has recommended that trainees complete the program once

256–258

(downgraded for imprecision) that time to respiration of every 2 years, but in the United Kingdom, 4 years is the recom-

more than 30 min was associated with 35% cerebral palsy and mended interval; there is no objective evidence to validate these

67–100% abnormal neurologic findings. intervals. It is intuitive that individual trainees will require different

For the critical outcome of death and/or moderate-to-severe training intervals to facilitate optimal acquisition and maintenance

disability, we identified very-low-quality evidence (downgraded of different skills. This PICO question is intended to update the

259,260

for imprecision) from 2 observational studies that time evidence as to what may be the most effective strategy.

to spontaneous respiration of 10–19 min and more than 20 min

was associated with this outcome in 56% and 88% of patients, Consensus on science

259

respectively, and time to spontaneous breathing of 30 min or Sixteen studies were identified that have investigated this PICO

261–270

more was a predictor of this outcome (OR, 2.33; 95% CI, 1.27–4.27). question. Ten randomized controlled studies and 6 nonran-

271–276

domized controlled trials were identified for inclusion.

Treatment recommendation The evidence for frequency of resuscitation training is very low

Absence of spontaneous breathing or an Apgar score of 1–3 at quality (downgraded for high risk of bias, inconsistency, and impre-

20 min of age in babies of greater than 34 weeks of gestation but cision), with the exception of studies of psychomotor performance,

with a detectable heart rate are strong predictors of mortality or which are of moderate quality (downgraded for risk of bias). Meta-

significant morbidity. In settings where resources are limited, we analyses were greatly limited by the heterogeneity between studies

suggest that it may be reasonable to stop assisted ventilation in of training frequency, educational interventions, and outcomes.

271,275

babies with no spontaneous breathing despite presence of heart For the critical outcome of patient outcomes, 2 studies of

rate or Apgar score of 1–3 at 20 min or more (weak recommenda- very low quality (downgraded for high risk of bias, inconsistency,

tion, very-low-quality evidence). and imprecision) looked at endotracheal intubation success. Both

studies included psychomotor skill training on an airway simu-

275

Values, preferences, and task force insights lator, and Nishisaki included simulation-based training. There

In making this statement, in infants of greater than 34 weeks was no significant difference in first-time intubation success (RR,

with an Apgar score of 0, 1, 2, or 3 for 20 min or more, the likelihood 0.879; 95% CI, 0.58–1.33) or any intubation success (RR, 0.87; 95%

of dying or having severe or moderate developmental disabilities CI, 0.65–1.17) between the providers who were exposed to frequent

at 18–24 months is very high. Importantly, each of the studies training and controls.

reviewed was conducted in a resource setting where therapeutic For the important outcome of prevention of adverse events, the

hypothermia was likely to be available (see NRP 734). Nishisaki study also included the important outcome of prevention

Perhaps there is a publication bias when those babies who of adverse outcomes and airway injury as a secondary outcome. No

did not respond at 20 min are not included in the numbers. The significant difference was seen between groups (RR, 1.097; 95% CI,

275

question was raised, if the prognosis is the same, why would we 0.747–1.612).

recommend something different for resource-limited settings? A For the important outcome of performance in simulation, 3

264,267,273

response was that in resource-limited regions, there will likely not studies of very low quality (downgraded for high risk

be the regional systems and postresuscitation neonatal intensive of bias, inconsistency, and imprecision) investigated the impor-

care facilities and subspecialty personnel that were available in tant outcome of performance in simulated scenarios using both

the recent studies reviewed in the Consensus on Science. If such validated and nonvalidated evaluations. In all studies, subjects

facilities are available, this treatment recommendation may be less in the intervention groups trained more frequently than con-

applicable. trols. The range of time between initial course completion and

first additional training session was 1–4 months. The educa-

tional interventions were heterogeneous, including independent Knowledge gaps

264

and facilitated practice on airway simulators, didactic lectures,

•

No studies identified from low-resource settings 273

skill station practice, mock codes, and periodic review of course

•

Outcome of babies with delayed onset of breathing who are man- 267 264 267

material and case-based study. Kovacs and Stross found

aged with induced hypothermia in low-resource settings.

no significant difference between frequent and infrequent practice

•

Outcome of babies with gasping or irregular breathing and a heart

with respect to simulation-based performance. Only 1 of these

activity at 20 min of life. 273

studies (Nadel ) offered quantitative data: After averaging of

multiple outcomes, there was a trend to improved performance

Educational techniques for teaching, assessing, and in those exposed to increased frequency of training compared with

maintaining resuscitation knowledge and skills controls (RR, 1.51; 95% CI, 0.971–2.35).

For the important outcome of psychomotor performance,

261,262,266,267,269,273,274,276

Resuscitation training frequency (NRP 859) there were 8 studies of moderate qual-

ity (downgraded for risk of bias) that evaluated the important

For course participants including (a) trainees and (b) practition- outcome of impact of frequent training on psychomotor perfor-

ers (P), does frequent training (I), compared with less frequent mance, demonstrated on a task trainer or simulator. With the

276 267

training (annual or biennial) (C), change all levels of education exception of O’Donnell and Stross (which were neutral to

or practice, prevention of adverse outcomes, overall mortality, the question), studies demonstrated improvements in psychomo-

scenario performance, medical knowledge, psychomotor perfor- tor performance with no negative effect. The range of time between

mance, provider confidence, course satisfaction (O)? course completion and first additional training session was 1 week

to 6 months. The educational interventions were again heteroge-

Introduction neous. Psychomotor task trainers were used to achieve competency

Training in the cognitive, technical, and behavioral skills nec- in a specific technical skill, including practice on a chest com-

274

essary for successful neonatal resuscitation has historically been pression task trainer (Niles ), neonatal airway management task

262

conducted at varying intervals of time, and there is little evidence trainer (Ernst ), or a CPR task trainer where both chest com-

261,266,276

to support the use of one interval over another. As an example, the pressions and ventilation were emphasized. The study

e192 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

267

by Stross included periodic review of course material and case- on studies, because the interventions themselves were so differ-

267 273

based study. The educational intervention in the Nadel study ent. Could the follow-up programs be briefer and more focused on

used didactic lectures, skill station practice, and mock codes. needs? What is best for the patient? What is the cost to the child

273

Although 8 studies were identified, only 1 randomized and 2 and family when the patient does not receive adequate resuscita-

267,276

observational studies with dichotomous quantitative data tion? What is a technical proficiency program? How do we achieve

273

were included in the analysis. The 1 randomized study demon- it? There is no assessment of translation of increased training to

strated a significant improvement in psychomotor skills in subjects improved outcomes. We need data to show that improved educa-

in the intervention group when compared with controls. One ran- tion is worth the staff time. The PICO question specifically avoided

266

domized study with multiple outcomes showed significantly looking at studies about decay of knowledge and skills.

improved performance of the important outcomes of manual ven-

tilation volume and chest compression depth after practice every

3 months. However, an improvement in psychomotor skills in the Knowledge gaps

267,273,276

intervention groups was not seen when 3 studies were •

Although some outcomes are of critical importance, the quality

included in a meta-analysis after averaging of scores (RR, 1.38; 95%

of evidence is very low. Serious methodological flaws occur, such

CI, 0.87–2.2).

as lack of randomization, multiple primary outcomes with inad-

For the important outcome of knowledge, 5

263,268,270,273,276 equate sample size and power analysis, lack of blinding, controls

studies of very low quality (downgraded for

that consist of no educational intervention resulting in a compar-

high risk of bias, inconsistency, and imprecision) investigated

ison of training to no training, insufficiently validated evaluation

the relationship between frequent training and the important

tools, and significant heterogeneity of outcomes and interven-

outcome of acquisition of medical knowledge assessed by writ-

273 260 tions.

ten tests or oral exams. Studies by Nadel, O’Donnell, and •

254 There is a need for well-designed and well-powered clinical tri-

Turner demonstrated sustained knowledge with refreshers

263 252 als, possibly cluster randomized, that answer key questions with

when compared with controls, whereas Kaczorowski and Su

critical outcomes: How frequently should learning occur? What

were neutral to the question. The educational interventions for

type of intervention is most effective? What validated tools are

these studies have been described previously except for 2 studies:

available to measure educational outcomes?

Su used a knowledge exam and mock resuscitation at 6 months, •

263 How do high-opportunity versus low-opportunity environments

and the Kaczorowski study included subjects in the intervention

differ in their need for frequent training?

groups either watching a newborn resuscitation education video or •

Did we take experience into account?

hands-on practice. The range of time between course completion •

What about knowledge, skills, and behaviors?

and first additional training session was 1–6 months. Although 5 •

270,273 Are patient outcomes lacking?

studies were identified, only 2 had quantitative data. The •

Is cost impact lacking?

analysis of the 2 observational studies was not possible because •

Is high-frequency, low-dose training effective?

it was difficult to average the means ± SDs and then pool the 2 •

273 Decay and boosting rates?

studies for a meta-analysis. The Nadel study found a significant •

Should we add “within the constraints of local resources”?

improvement in knowledge with more frequent training in a short •

Reinforcement from other domains, for example

answer test (mean scores 73 ± 11 versus 60 ± 10; P = 0.0003). The

254

Turner study showed significant improvement in 2 out of 3 test

scores in the intervention group (mean scores 7.1 versus 6.2 and Neonatal resuscitation instructors (NRP 867)

260

29.0 versus 25.8, respectively; P < 0.05 in both cases). O’Donnell

demonstrated lower test scores in the control group than in the In neonatal resuscitation instructors (P), does formal train-

intervention group (P < 0.04). ing on specific aspects of how to facilitate learning (I), compared

For the nonimportant outcome of provider confidence, with generic or nonspecific training (C), change clinical outcome,

265

Montgomery found that subjects who practiced CPR for 6 min improve all levels of education or practice (O)?

every month were more likely than controls to report that they

273

felt confident (RR, 1.60; 95% CI, 1.27–2.01), and Nadel found

improved confidence in both leadership and technical skills. Introduction

No study demonstrated a negative or detrimental effect from Around the world, millions of healthcare professionals bear

more frequent training. Publication bias was difficult to assess. the responsibility for resuscitating neonates in the delivery room,

and they must not only acquire the necessary cognitive, techni-

cal, and behavioral skills but also maintain them over time, often

Treatment recommendation

for decades. The precise roles and mandatory skills of the instruc-

We suggest that training should be recurrent and considered

tors charged with training healthcare professionals have yet to be

more frequently than once per year. This retraining may be com-

defined, and thus how to best prepare instructors to fulfill these

posed of specific tasks and/or behavioral skills, depending on the

roles and acquire these skills is not yet objectively described. It is

needs of the trainees (weak recommendation, very-low-quality

intuitive that training of instructors should be based on specific

evidence).

learning objectives targeting the specific instructor skills that are

necessary to facilitate the acquisition of specific skills in specific

Values, preferences, and task force insights populations of learners. Comprehensive assessment of resuscita-

In drawing our conclusions, we place value on improved tion instructor training requires identification and development

psychomotor skills, knowledge, and provider confidence during of (1) objective markers of performance for instructors, (2) appro-

more-frequent training versus less-frequent training (and versus priate objective markers of performance for the trainees who are

the established and unproven practice of training every 1–2 years). trained by the instructors, and (3) objective markers of patient out-

The debate included the fact that the PICO question does not come that are directly related to how well they were resuscitated.

specify that it is resuscitation training, although the search did This PICO question is intended to identify literature that is pertinent

restrict itself to this. Should the costs of training be addressed? to these and other issues involving the preparation of instructor of

However, it was noted that it was hard to comment on cost based neonatal resuscitation.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e193

Consensus on science factors and ergonomics who have experience in examining human

For the critical outcome of improvement in patient outcome, performance in high-risk domains (similar to the delivery room)

we identified no evidence. rather than relying solely on those with expertise in traditional

For the critical outcome of improvement in learner perfor- education settings such as the classroom.

mance in the real clinical environment, we identified very-low-

277

quality evidence from 1 randomized clinical trial (downgraded Deliberations of the task force and writing group

for indirectness, risk of bias, and imprecision) that providing The PICO question may be too global/broad. Perhaps we need to

structured self-reflection and peer group feedback to psychiatry be more specific in the future. We may need to move away from

registrars improved their students’ performance of standardized dependence on traditional methodologies and look to those indus-

psychiatric interviews. tries where adults are trained to be proficient in specific tasks.

For the critical outcome of improvement in learner perfor- Instructors need to know how to do specific tasks and give feed-

mance in educational settings, we identified very-low-quality back to improve performance. Perhaps we have made instructors

evidence (downgraded for indirectness, imprecision, and risk of poor trainers. People who develop curricula need to address this

278

bias) from 1 randomized clinical trial in which 18 emergency critical deficit. How do we teach task proficiency? That is what is

medicine instructors were randomly assigned to 2 intervention most needed.

groups and trained 193 medical students. The study found that

learners trained by instructors who underwent a 2-day teacher Knowledge gaps

training course focused on education principles performed at an •

How is optimal instructor performance defined?

equal or lower level of proficiency in technical skills when com- •

What are the skills necessary to achieve this?

pared with those trained by instructors who did not attend the •

What are the optimal methods for selection of candidate instruc-

2-day course.

tors, initial skill acquisition by instructors, ongoing maintenance

For the critical outcome of improvement in all levels of educa-

of instructor skill, and (objective and subjective) assessment of

tion or practice, we identified low-quality evidence (downgraded

278–282 instructor skill?

for indirectness and bias) from 5 randomized clinical trial

enrolling 271 participants (not estimable). Several studies did note

2010 PICO questions not reviewed in 2015

at least temporary deterioration in instructor performance after

commencement of new instructor training intervention. •

Suctioning (other than meconium)

For the critical outcome of improvement in clinical outcome, •

Inflation pressures

we identified no evidence. •

Face mask characteristics

For the important outcome of improvement in instructor per- •

CO2 detectors to confirm endotracheal tube placement

formance, we identified very-low-quality evidence (downgraded •

278–282 Epinephrine dose and route

for indirectness and bias) from 5 randomized clinical trials •

283,284 Volume expansion

and 2 nonrandomized trials. No meaningful numerical sum- •

Sodium bicarbonate

mary of the results of these studies could be performed. These

• Glucose

studies indicate that preparation of instructors produces inconsis- •

Therapeutic hypothermia

tent results in terms of instructor performance. While it does seem •

Personnel needs at elective cesarean delivery

that written and verbal feedback, delivered in a constructive and •

Briefing and debriefings during learning activities

timely manner, often produces improvement in instructor perfor-

mance, in other instances posttraining deterioration in aspects of

Acknowledgments

instructor performance was seen, at least initially.

We thank the following evidence reviewers (the Neonatal

Treatment recommendation

Resuscitation Chapter Collaborators) for their great effort, due dili-

We suggest that training of resuscitation instructors incorporate

gence, and expertise with regard to the reviews contained in this

timely, objective, structured, individually targeted verbal and/or

section: David W. Boyle, Steve Byrne, Chris Colby, Peter Davis, Hege

written feedback (weak recommendation, low-quality evidence).

L. Ersdal, Marilyn B. Escobedo, Qi Feng, Maria Fernanda de Almeida,

Louis P. Halamek, Tetsuya Isayama, Vishal S. Kapadia, Henry C. Lee,

Values, preferences, and task force insights

Jane McGowan, Douglas D. McMillan, Susan Niermeyer, Colm P.

While common sense dictates that instructors be properly pre-

F. O’Donnell, Yacov Rabi, Steven A. Ringer, Nalini Singhal, Ben J.

pared before engaging learners, it is clear that such instruction must

Stenson, Marya L. Strand, Takahiro Sugiura, Daniele Trevisanuto,

be based on specific learning objectives targeting the specific skills

Enrique Udaeta, Gary M. Weiner, and Cheo L. Yeo. We also acknowl-

that are necessary to facilitate learning. Definitions of these skills

edge the comments received during the public period.

will require collaboration with colleagues in fields such as human

e194 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Disclosures. 2015 CoSTR Part 7: neonatal resuscitation: writing group

Writing group Employment Research Other research Speakers’ Expert Ownership Consultant/advisory Other

member grant support bureau/honoraria witness interest board

Jonathan Wyllie James Cook None None None None None None None

University Hospital

Jeffrey M. Perlman Weill Cornell None None None None None None None

Medical College

Khalid Aziz University of None None None None None None None

Alberta Pediatrics

Ruth Guinsburg Federal University None None None None None None None

of Sao Paulo

John Kattwinkel University of None None None None None None None

Virginia Health

System

Han-Suk Kim Seoul National None None None None None None None

University College

of Medicine

Helen G. Liley Mater Mother’s NHMRC None None None None None Mater

Hospital (Australia)a Foundationb

Lindsay Mildenhall Middlemore None None None None None None None Hospital

b

Wendy M. Simon American Academy None None None None None None AAP staff

of Pediatrics

Edgardo Szyld University of None None None None None None None

Oklahoma

Masanori Tamura Saitama Medical None None None None None None None

Center

Sithembiso Velaphi Resuscitation None None None None None None None

Council of

Southern Africa

Myra H. Wyckoff UT Southwestern None None None None None None None

This table represents the relationships of writing group members that may be perceived as actual or reasonably perceived conflicts of interest as reported on the Disclosure

Questionnaire, which all members of the writing group are required to complete and submit. A relationship is considered to be “significant” if (a) the person receives $10,000

or more during any 12-month period, or 5% or more of the person’s gross income; or (b) the person owns 5% or more of the voting stock or share of the entity, or owns

$10,000 or more of the fair market value of the entity. A relationship is considered to be “modest” if it is less than “significant” under the preceding definition. a Modest.

b Significant.

Appendix A. CoSTR Part 7: PICO Appendix

Part Task force PICO ID Short title PICO question Evidence reviewers

Part 7 NRP NRP 589 Temperature maintenance in In nonasphyxiated babies at birth (P), does maintenance of Jonathan Wyllie,

◦ ◦

the delivery room—prognosis normothermia (core temperature 36.5 C or greater and 37.5 C or Jeffrey Perlman

less) from delivery to admission (I), compared with hypothermia

◦ ◦

(less than 36 C) or hyperthermia (greater than 37.5 C) (C), change

survival to hospital discharge, respiratory distress, survival to

admission, hypoglycemia, intracranial hemorrhage, or infection

rate (O)?

Part 7 NRP NRP 590 CPAP and IPPV—intervention In spontaneously breathing preterm infants with respiratory Tetsuya Isayama,

distress requiring respiratory support in the delivery room (P), Ben Stenson

does the use of CPAP (I), compared with intubation and IPPV (C),

improve outcome (O)?

Part 7 NRP NRP 599 Maintaining infant Among preterm neonates who are under radiant warmers in the Daniele

temperature during delivery hospital delivery room (P), does increased room temperature, Trevisanuto, Maria

room thermal mattress, or another intervention (I), compared with Fernanda de

◦

resuscitation—intervention plastic wraps alone (C), reduce hypothermia (less than 36 C) on Almeida

admission to neonatal intensive care unit (NICU) (O)?

Part 7 NRP NRP 605 Thumb versus 2-finger In neonates receiving cardiac compressions (P), does the use of a Myra Wyckoff,

techniques for chest 2-thumb technique (I), compared with a 2-finger technique (C), Lindsay Mildenhall

compression—intervention result in return of spontaneous circulation (ROSC), improved

neurologic outcomes, improved survival, improved perfusion and

gas exchange during CPR, and decreased compressor fatigue (O)?

Part 7 NRP NRP 618 Laryngeal mask In newborn infants at near term (greater than 34 weeks) or term Edgardo Szyld,

airway—intervention who have indications for intermittent positive pressure for Enrique Udaeta

resuscitation (P), does use of a laryngeal mask as a primary or

secondary device (I), compared with mask ventilation or

endotracheal intubation (C), improve response to resuscitation or

change outcome (O), including indicators of neonatal brain injury,

achieving stable vital signs, increasing Apgar scores, long-term

outcomes, reducing the need for subsequent intubation, or

neonatal morbidity and mortality?

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e195

Part Task force PICO ID Short title PICO question Evidence reviewers

Part 7 NRP NRP 734 Limited-resource–induced In term infants with moderate/severe hypoxic–ischemic Peter Davis, Jeffrey

hypothermia—intervention encephalopathy managed in resource-limited countries (P), does Perlman

therapeutic hypothermia to core temperature of approximately

◦

33.5 C for 72 h delivered by passive hypothermia and/or ice packs

(I), versus standard therapy (C), improve the rates of death,

neurodevelopmental impairments at 18 months to 2 years (O)?

Part 7 NRP NRP 738 Oxygen delivery during CPR In neonates receiving cardiac compressions (P), does 100% O2 as Myra Wyckoff,

(neonatal)—intervention the ventilation gas (I), compared with lower concentrations of Lindsay Mildenhall

oxygen (C), increase survival rates, improve neurologic outcomes,

decrease time to ROSC, or decrease oxidative injury (O)?

Part 7 NRP NRP 787 Delayed cord clamping in In preterm infants, including those who received resuscitation (P), Masanori Tamura,

preterm infants requiring does delayed cord clamping (greater than 30 seconds) (I), Susan Niermeyer

resuscitation (intervention) compared with immediate cord clamping (C), improve survival,

long-term developmental outcome, cardiovascular stability,

occurrence of intraventricular hemorrhage (IVH), necrotizing

enterocolitis, temperature on admission to a newborn area, and

hyperbilirubinemia (O)?

Part 7 NRP NRP 793 Maintaining infant In newborn infants (greater than 30 weeks of gestation) in Sithembiso

temperature during delivery low-resource settings during and/or after Velaphi, Hege

room resuscitation/stabilization (P), does drying and skin-to-skin Ersdal, Nalini

resuscitation—intervention contact or covering with plastic (I), compared with drying and no Singhal

skin-to-skin or use of radiant warmer or incubator (C), change

body temperature (O)?

Part 7 NRP NRP 804 Babies born to mothers who In newborn babies (P), does maternal hypothermia or Henry Lee, Marilyn

are hypothermic or hyperthermia in labor (I), versus normal maternal temperature (C), Escobedo

hyperthermic in result in adverse neonatal effects (O)? Outcomes include mortality,

labor—prognosis neonatal seizures, and adverse neurologic states

Part 7 NRP NRP 805 Delivery room assessment for In extremely preterm infants (less than 25 weeks) (P), does Steven Ringer,

less than 25 weeks and delivery room assessment with a prognostic score (I), compared Steve Byrne

prognostic score with gestational age assessment alone (C), change survival to 18 to

22 months (O)?

Part 7 NRP NRP 806 Newborn infants who receive In newborn infants who receive PPV for resuscitation (P), does use Helen Liley, Vishal

PPV for resuscitation, and use of a device to assess respiratory function with or without pressure Kapadia

of a device to assess respiratory monitoring (I), compared with no device (C), change survival to

function—diagnostic hospital discharge with good neurologic outcome, IVH, time to

heart rate greater than 100/min, bronchopulmonary dysplasia,

pneumothorax (O)?

Part 7 NRP NRP 809 Sustained In term and preterm newborn infants who do not establish Jane McGowan,

inflations—intervention spontaneous respiration at birth (P), does administration of 1 or David Boyle

more pressure-limited sustained lung inflations (I), compared with

intermittent PPV with short inspiratory times (C), change Apgar

score at 5 min, establishment of FRC, requirement for mechanical

ventilation in first 72 h, time to heart rate greater than 100/min,

rate of tracheal intubation, overall mortality (O)?

Part 7 NRP NRP 849 Umbilical cord In very preterm infants (28 weeks or less) (P), does umbilical cord Marya Strand,

milking—intervention milking (I), in comparison with immediate umbilical cord Takahiro Sugiura

clamping (C), affect death, neurodevelopmental outcome at 2–3

years, cardiovascular stability, i.e. need for pressors, need for fluid

bolus, initial mean blood pressure, IVH (any grade, severe grade),

temperature on admission, hematologic indices (initial

hemoglobin, need for transfusion), hyperbilirubinemia, need for

phototherapy, or need for exchange transfusion (O)?

◦

Part 7 NRP NRP 858 Warming of hypothermic In newborns who are hypothermic (temperature less than 36.0 C) Cheo Yeo, Daniele

newborns—intervention on admission (P), does rapid rewarming (I), compared with slow Trevisanuto

rewarming (C), change mortality rate, short and long-term

neurologic outcome, hemorrhage, episodes of apnea and

hypoglycemia, or need for respiratory support (O)?

Part 7 NRP NRP 859 Resuscitation training For course participants including (a) trainees and (b) practitioners Chris Colby, Khalid

frequency (P), does frequent training (I), compared with less frequent Aziz

training (annual or biennial) (C), change all levels of education or

practice, prevention of adverse outcomes, overall mortality,

scenario performance, medical knowledge, psychomotor

performance, provider confidence, course satisfaction (O)?

Part 7 NRP NRP 860 Predicting death or disability of In newborn infants of greater than 34 weeks of gestation, receiving Sithembiso

newborns of greater than 34 PPV at birth in settings where resources are limited (P), does Velaphi, Nalini

weeks based on Apgar and/or presence of heart rate with no spontaneous breathing or Apgar Singhal, Hege

absence of scores of 1–3 at greater than 5 min predict mortality or morbidity Ersdal

breathing—prognosis or cerebral palsy (O)?

Part 7 NRP NRP 862 Use of feedback CPR devices for In asystolic/bradycardic neonates receiving cardiac compressions Lindsay Mildenhall,

neonatal cardiac (P), does the use of feedback devices such as end-tidal carbon Takahiro Sugiura

arrest—diagnostic dioxide (ETCO2) monitors, pulse oximeters, or automated

compression feedback devices (I), compared with clinical

assessments of compression efficacy (C), decrease hands-off time,

decrease time to ROSC, improve perfusion, increase survival rates,

or improve neurologic outcomes (O)?

e196 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

Part Task force PICO ID Short title PICO question Evidence reviewers

Part 7 NRP NRP 864 Oxygen concentration for Among preterm newborns (less than 37 weeks of gestation) Gary Weiner,

resuscitating premature who receive PPV in the delivery room (P), does the use of high Douglas McMillan

newborns—intervention O2 (50–100%) as the ventilation gas (I), compared with low

concentrations of O2 (21–30%) (C), decrease mortality,

decrease bronchopulmonary dysplasia, decrease retinopathy,

decrease IVH (O)?

Part 7 NRP NRP 865 Intubation and tracheal In nonvigorous infants at birth born through MSAF (P), does Sithembiso

suctioning in nonvigorous tracheal intubation for suctioning (I), compared with no Velaphi, Jeffrey

infants born through MSAF tracheal intubation (C), reduce meconium syndrome or Perlman

versus no intubation for prevent death (O)? tracheal

suctioning—intervention

Part 7 NRP NRP 867 Neonatal resuscitation In neonatal resuscitation instructors (P), does formal training Helen Liley, Louis

instructors on specific aspects of how to facilitate learning (I), compared Halamek

with generic or nonspecific training (C), change clinical

outcome, improve all levels of education or practice (O)?

Part 7 NRP NRP 870 T-piece resuscitator and In newborns (preterm and term) receiving ventilation (PPV) Yacov Rabi, Han

self-inflating bag—intervention during resuscitation (P), does using a T-piece resuscitator with Suk Kim

PEEP (I), compared with using a self-inflating bag without

PEEP (C), achieve spontaneous breathing sooner and/or reduce

the incidence of pneumothorax, bronchopulmonary dysplasia,

and mortality (O)?

Part 7 NRP NRP 895 Chest compression In neonates receiving cardiac compressions (P), do other ratios Qi Feng, Myra

ratio—intervention (5:1, 9:3, 15:2, synchronous, etc) (I), compared with 3:1 Wyckoff

compressions to ventilations (C), increase survival rates,

improve neurologic outcomes, improve perfusion and gas

exchange during CPR, decrease time to ROSC, decrease tissue

injury, or decrease compressor fatigue (O)?

Part 7 NRP NRP 896 Apgar score of 0 for 10 min or In infants with a gestational age of 36 weeks or greater and an Ruth Guinsburg,

longer—prognosis Apgar score of 0 for 10 min or longer, despite ongoing Jane McGowan

resuscitation (P), what is the rate of survival to NICU admission

and death or neurocognitive impairment at 18–22 months (O)?

Part 7 NRP NRP 897 Outcomes for PEEP versus no In preterm/term newborn infants who do not establish Yacov Rabi, Colm

PEEP in the delivery respiration at birth (P), does the use of PEEP as part of the O’Donnell

room—intervention initial ventilation strategy (I), compared with no PEEP (C),

improve Apgar score at 5 min, intubation in the delivery room,

chest compressions in the delivery room, heart rate greater

than 100/min by 2 min of life, time for heart rate to rise above

100/min, air leaks, oxygen saturation/oxygenation, Fio2 in the

delivery room, mechanical ventilation in the first 72 h,

bronchopulmonary dysplasia, survival to discharge (O)?

Part 7 NRP NRP 898 ECG/EKG (I) in comparison to In babies requiring resuscitation (P), does electrocardiography Marya Strand, Hege

oximetry or auscultation for (ECG/EKG) (I), compared with oximetry or auscultation (C), Ersdal

the detection of heart rate measure heart rate faster and more accurately (O)?

11. Schünemann H, Brozek˙ J, Guyatt G, Oxman A. 5. Quality of evidence. In:

References GRADE handbook; 2013. http://www.guidelinedevelopment.org/handbook/

#h.9rdbelsnu4iy [accessed 06.05.15].

1. Ersdal HL, Mduma E, Svensen E, Perlman JM. Early initiation of basic resuscita- 12. Schünemann H, Brozek˙ J, Guyatt G, Oxman A. 5.1 Factors determin-

tion interventions including face mask ventilation may reduce birth asphyxia ing the quality of evidence. In: GRADE handbook; 2013. http://www.

related mortality in low-income countries: a prospective descriptive observa- guidelinedevelopment.org/handbook/#h.9rdbelsnu4iy [accessed 06.05.15].

tional study. Resuscitation 2012;83:869–73. 13. Perlman JM, Wyllie J, Kattwinkel J, et al., Neonatal Resuscitation Chapter

2. Perlman JM, Risser R. Cardiopulmonary resuscitation in the delivery room. Collaborators. Part 11: neonatal resuscitation: 2010 International Con-

Associated clinical events. Arch Pediatr Adolesc Med 1995;149:20–5. sensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular

3. Barber CA, Wyckoff MH. Use and efficacy of endotracheal versus intravenous Care Science With Treatment Recommendations. Circulation 2010;122:

epinephrine during neonatal cardiopulmonary resuscitation in the delivery S516–38.

room. Pediatrics 2006;118:1028–34. 14. Perlman JM, Wyllie J, Kattwinkel J, et al. Neonatal resuscitation: 2010

4. Institute of Medicine. Standards for Systematic Reviews; 2011. http://www. International Consensus on Cardiopulmonary Resuscitation and Emergency

iom.edu/Reports/2011/Finding-What-Works-in-Health-Care-Standards-for- Cardiovascular Care Science with Treatment Recommendations. Pediatrics

Systematic-Reviews/Standards.aspx [accessed 06.05.15]. 2010;126:e1319–44.

5. Schünemann H, Brozek˙ J, Guyatt G, Oxman A. GRADE handbook; 2013. http:// 15. Wyllie J, Perlman JM, Kattwinkel J, et al. Part 11: neonatal resuscitation: 2010

www.guidelinedevelopment.org/handbook/ [accessed 06.05.15]. International Consensus on Cardiopulmonary Resuscitation and Emergency

6. Chapter 5: Defining the review questions and developing criteria for including Cardiovascular Care Science with Treatment Recommendations. Resuscitation

studies.O’Connor D, Green S, Higgins JPT, The Cochrane Collaboration, Higgins J, 2010;81:e260–87.

Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. 16. Perlman J, Kattwinkel J, Wyllie J, Guinsburg R, Velaphi S, Singhal N, for the

Version 5.1.0. 2011. http://handbook.cochrane.org/ [accessed 06.05.15]. Neonatal ILCOR and Task Force Group. Neonatal resuscitation: in pursuit of

7. Chapter 8.5: The Cochrane Collaboration’s tool for assessing risk of bias.Higgins evidence gaps in knowledge. Resuscitation 2012;83:545–50.

JPT, Altman DG, Sterne J, The Cochrane Collaboration, Higgins JPT, Green S, 17. Chettri S, Adhisivam B, Bhat BV. Endotracheal suction for nonvigorous neonates

editors. Cochrane Handbook for Systematic Reviews of Interventions. Version born through meconium stained amniotic fluid: a randomized controlled trial.

5.1.0. 2011. http://handbook.cochrane.org/ [accessed 06.05.15]. J Pediatr 2015;166:1208.e1–5.

8. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a revised tool 18. Lista G, Boni L, Scopesi F, et al. Sustained lung inflation at birth for preterm

for the quality assessment of diagnostic accuracy studies. Ann Intern Med infants: a randomized clinical trial. Pediatrics 2015;135:e457–64.

2011;155:529–36. 19. Dawson JA, Saraswat A, Simionato L, et al. Comparison of heart rate and oxygen

9. Schünemann H, Brozek˙ J, Guyatt G, Oxman A. 5.2.1 Study limitations (risk saturation measurements from Masimo and Nellcor pulse oximeters in newly

of bias). In: GRADE Handbook; 2013. http://www.guidelinedevelopment.org/ born term infants. Acta Paediatr 2013;102:955–60.

handbook/#h.m9385o5z3li7 [accessed 06.05.15]. 20. Kamlin CO, Dawson JA, O’Donnell CP, et al. Accuracy of pulse oximetry mea-

10. Evidence Prime Inc. GRADEpro Guideline Development Tool. http://www. surement of heart rate of newborn infants in the delivery room. J Pediatr

guidelinedevelopment.org/ [accessed 06.05.15]. 2008;152:756–60.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e197

21. Katheria A, Rich W, Finer N. Electrocardiogram provides a continuous 50. Costeloe K, Hennessy E, Gibson AT, Marlow N, Wilkinson AR. The EPICure

heart rate faster than oximetry during neonatal resuscitation. Pediatrics study: outcomes to discharge from hospital for infants born at the threshold

2012;130:e1177–81. of viability. Pediatrics 2000;106:659–71.

22. Mizumoto H, Tomotaki S, Shibata H, et al. Electrocardiogram shows reliable 51. Costeloe KL, Hennessy EM, Haider S, Stacey F, Marlow N, Draper ES. Short term

heart rates much earlier than pulse oximetry during neonatal resuscitation. outcomes after extreme preterm birth in England: comparison of two birth

Pediatr Int 2012;54:205–7. cohorts in 1995 and 2006 (the EPICure studies). BMJ 2012;345:e7976.

23. van Vonderen JJ, Hooper SB, Kroese JK, et al. Pulse oximetry measures 52. da Mota Silveira SM, Gonc¸ alves de Mello MJ, de Arruda Vidal S, de Frias PG,

a lower heart rate at birth compared with electrocardiography. J Pediatr Cattaneo A. Hypothermia on admission: a risk factor for death in newborns

2015;166:49–53. referred to the Pernambuco Institute of Mother and Child Health. J Trop Pediatr

24. Kamlin CO, O’Donnell CP, Everest NJ, Davis PG, Morley CJ. Accuracy of clin- 2003;49:115–20.

ical assessment of infant heart rate in the delivery room. Resuscitation 53. Daga AS, Daga SR, Patole SK. Determinants of death among admissions to

2006;71:319–21. intensive care unit for newborns. J Trop Pediatr 1991;37:53–6.

25. Baenziger O, Stolkin F, Keel M, et al. The influence of the timing of cord clam- 54. de Almeida MF, Guinsburg R, Sancho GA, et al. Hypothermia and early neonatal

ping on postnatal cerebral oxygenation in preterm neonates: a randomized, mortality in preterm infants. J Pediatr 2014;164:271.e1–5.

controlled trial. Pediatrics 2007;119:455–9. 55. García-Munoz˜ Rodrigo F, Rivero Rodríguez S, Siles Quesada C. Hypothermia risk

26. Hofmeyr GJ, Bolton KD, Bowen DC, Govan JJ. Periventricular/intraventricular factors in the very low weight newborn and associated morbidity and mortality

haemorrhage and umbilical cord clamping. Findings and hypothesis. S Afr Med in a neonatal care unit. An Pediatr (Barc) 2014;80:144–50.

J 1988;73:104–6. 56. Harms K, Osmers R, Kron M, et al. Mortality of premature infants 1980–1990:

27. Hofmeyr GJ, Gobetz L, Bex PJ, et al. Periventricular/intraventricular hemorrhage analysis of data from the Göttingen Perinatal Center. Z Geburtshilfe Perinatol

following early and delayed umbilical cord clamping. A randomized controlled 1994;198:126–33.

trial. Online J Curr Clin Trials 1993. Doc No 110: [2002 words; 26 paragraphs]. 57. Hazan J, Maag U, Chessex P. Association between hypothermia and mortality

28. Kinmond S, Aitchison TC, Holland BM, Jones JG, Turner TL, Wardrop CA. rate of premature infants-revisited. Am J Obstet Gynecol 1991;164:111–2.

Umbilical cord clamping and preterm infants: a randomised trial. BMJ 58. Jones P, Alberti C, Julé L, et al. Mortality in out-of-hospital premature births.

1993;306:172–5. Acta Paediatr 2011;100:181–7.

29. Kugelman A, Borenstein-Levin L, Riskin A, et al. Immediate versus delayed 59. Kambarami R, Chidede O. Neonatal hypothermia levels and risk factors for

umbilical cord clamping in premature neonates born <35 weeks: a prospective, mortality in a tropical country. Cent Afr J Med 2003;49:103–6.

randomized, controlled study. Am J Perinatol 2007;24:307–15. 60. Kent AL, Williams J. Increasing ambient operating theatre temperature and

30. McDonnell M, Henderson-Smart DJ. Delayed umbilical cord clamping in wrapping in polyethylene improves admission temperature in premature

preterm infants: a feasibility study. J Paediatr Child Health 1997;33: infants. J Paediatr Child Health 2008;44:325–31.

308–10. 61. Lee HC, Ho QT, Rhine WD. A quality improvement project to improve admission

31. Mercer JS, McGrath MM, Hensman A, Silver H, Oh W. Immediate and delayed temperatures in very low birth weight infants. J Perinatol 2008;28:754–8.

cord clamping in infants born between 24 and 32 weeks: a pilot randomized 62. Levi S, Taylor W, Robinson LE, Levy LI. Analysis of morbidity and outcome of

controlled trial. J Perinatol 2003;23:466–72. infants weighing less than 800 grams at birth. South Med J 1984;77:975–8.

32. Mercer JS, Vohr BR, McGrath MM, Padbury JF, Wallach M, Oh W. Delayed cord 63. Manani M, Jegatheesan P, DeSandre G, Song D, Showalter L, Govindaswami

clamping in very preterm infants reduces the incidence of intraventricular B. Elimination of admission hypothermia in preterm very low-birth-

hemorrhage and late-onset sepsis: a randomized, controlled trial. Pediatrics weight infants by standardization of delivery room management. Perm J

2006;117:1235–42. 2013;17:8–13.

33. Oh W, Fanaroff AA, Carlo WA, et al. Effects of delayed cord clamping in very- 64. Manji KP, Kisenge R. Neonatal hypothermia on admission to a special care unit

low-birth-weight infants. J Perinatol 2011;31:S68–71. in Dar-es-Salaam. Tanzania: a cause for concern. Cent Afr J Med 2003;49:23–7.

34. Rabe H, Wacker A, Hülskamp G, et al. A randomised controlled trial of 65. Mathur NB, Krishnamurthy S, Mishra TK. Evaluation of WHO classification of

delayed cord clamping in very low birth weight preterm infants. Eur J Pediatr hypothermia in sick extramural neonates as predictor of fatality. J Trop Pediatr

2000;159:775–7. 2005;51:341–5.

35. Strauss RG, Mock DM, Johnson KJ, et al. A randomized clinical trial comparing 66. Miller SS, Lee HC, Gould JB. Hypothermia in very low birth weight infants:

immediate versus delayed clamping of the umbilical cord in preterm infants: distribution, risk factors and outcomes. J Perinatol 2011;31:S49–56.

short-term clinical and laboratory endpoints. Transfusion 2008;48:658–65. 67. Zayeri F, Kazemnejad A, Ganjali M, Babaei G, Khanafshar N, Nayeri F. Hypother-

36. Aladangady N, McHugh S, Aitchison TC, Wardrop CA, Holland BM. Infants’ mia in Iranian newborns, Incidence, risk factors and related complications.

blood volume in a controlled trial of placental transfusion at preterm delivery. Saudi Med J 2005;26:1367–71.

Pediatrics 2006;117:93–8. 68. Obladen M, Heemann U, Hennecke KH, Hanssler L. Causes of neonatal mortality

37. Committee Opinion No. 543: timing of umbilical cord clamping after birth. 1981–1983: a regional analysis. Z Geburtshilfe Perinatol 1985;189:181–7.

Obstet Gynecol 2012;120:1522–6. 69. Ogunlesi TA, Ogunfowora OB, Adekanmbi FA, Fetuga BM, Olanrewaju DM.

38. Hosono S, Mugishima H, Fujita H, et al. Umbilical cord milking reduces the need Point-of-admission hypothermia among high-risk Nigerian newborns. BMC

for red cell transfusions and improves neonatal adaptation in infants born at Pediatr 2008;8:40.

less than 29 weeks’ gestation: a randomised controlled trial. Arch Dis Child 70. Pal DK, Manandhar DS, Rajbhandari S, et al. Neonatal hypoglycaemia in Nepal 1.

Fetal Neonatal Ed 2008;93:F14–9. Prevalence and risk factors. Arch Dis Child Fetal Neonatal Ed 2000;82:F46–51.

39. Katheria AC, Leone TA, Woelkers D, Garey DM, Rich W, Finer NN. The effects of 71. Shah S, Zemichael O, Meng HD. Factors associated with mortality and length

umbilical cord milking on hemodynamics and neonatal outcomes in premature of stay in hospitalised neonates in Eritrea, Africa: a cross-sectional study. BMJ

neonates. J Pediatr 2014;164:1045. Open 2012;2, http://dx.doi.org/10.1136/bmjopen-2011-000792.

40. March MI, Hacker MR, Parson AW, Modest AM, de Veciana M. The effects of

72. Singh A, Yadav A, Singh A. Utilization of postnatal care for newborns and its

umbilical cord milking in extremely preterm infants: a randomized controlled

association with neonatal mortality in India: an analytical appraisal. BMC Preg-

trial. J Perinatol 2013;33:763–7.

nancy Childbirth 2012;12:33.

41. Budin, P. The nursling. The feeding and hygiene of premature and full-term

73. Sodemann M, Nielsen J, Veirum J, Jakobsen MS, Biai S, Aaby P. Hypothermia of

infants. Translation by Maloney, WJ. Lecture 1. Page 1 London: The Caxton

newborns is associated with excess mortality in the first 2 months of life in

Publishing Company; 1907.

Guinea-Bissau, West Africa. Trop Med Int Health 2008;13:980–6.

42. Mullany LC, Katz J, Khatry SK, LeClerq SC, Darmstadt GL, Tielsch JM. Risk of mor-

74. Stanley FJ, Alberman EV. Infants of very low birthweight. I: Perinatal factors

tality associated with neonatal hypothermia in southern Nepal. Arch Pediatr

affecting survival. Dev Med Child Neurol 1978;20:300–12.

Adolesc Med 2010;164:650–6.

75. Wyckoff MH, Perlman JM. Effective ventilation and temperature control are

43. Laptook AR, Salhab W, Bhaskar B, Neonatal Research Network. Admission tem-

vital to outborn resuscitation. Prehosp Emerg Care 2004;8:191–5.

perature of low birth weight infants: predictors and associated morbidities.

76. Kalimba EM, Ballot D. Survival of extremely low-birth-weight infants. SAJCH

Pediatrics 2007;119:e643–9. 2013;7:13–6.

44. Abd-El Hamid A, Badr-El Din S, Dabous MM, Saad NIKM. Effect of the use of

77. Nayeri FFN. Hypothermia at birth and its associated complications in new-

a polyethylene wrap on the morbidity and mortality of very low birth weight

borns: a follow-up study. Iran J Public Health 2006;35:48–52.

infants in Alexandria University Children’s Hospital. J Egypt Public Health Assoc

78. Meyer MP, Payton MJ, Salmon A, Hutchinson C, de Klerk A. A clinical comparison

2012;87:104–8.

of radiant warmer and incubator care for preterm infants from birth to 1800

45. Acolet D, Elbourne D, McIntosh N, et al. Project 27/28: inquiry into quality of

grams. Pediatrics 2001;108:395–401.

neonatal care and its effect on the survival of infants who were born at 27 and 28

79. Reilly MC, Vohra S, Rac VE, et al. Randomized trial of occlusive wrap for heat

weeks in England, Wales, and Northern Ireland. Pediatrics 2005;116:1457–65.

loss prevention in preterm infants. J Pediatr 2015;166:262-8.e2.

46. Bateman DA, O’Bryan L, Nicholas SW, Heagarty MC. Outcome of unattended

80. Vohra S, Frent G, Campbell V, Abbott M, Whyte R. Effect of polyethylene occlu-

out-of-hospital births in Harlem. Arch Pediatr Adolesc Med 1994;148:147–52.

sive skin wrapping on heat loss in very low birth weight infants at delivery: a

47. Bhoopalam PS, Watkinson M. Babies born before arrival at hospital. Br J Obstet

randomized trial. J Pediatr 1999;134:547–51.

Gynaecol 1991;98:57–64.

81. Vohra S, Roberts RS, Zhang B, Janes M, Schmidt B. Heat Loss Prevention (HeLP)

48. Boo NY, Guat-Sim Cheah I, Malaysian National Neonatal Registry. Admis-

in the delivery room: a randomized controlled trial of polyethylene occlusive

sion hypothermia among VLBW infants in Malaysian NICUs. J Trop Pediatr

skin wrapping in very preterm infants. J Pediatr 2004;145:750–3.

2013;59:447–52.

82. Billimoria Z, Chawla S, Bajaj M, Natarajan G. Improving admission tempera-

49. Buetow KC, Kelein SW. Effects of maintenance of “normal” skin temperature

ture in extremely low birth weight infants: a hospital-based multi-intervention

on survival of infants of low birth weight. Pediatrics 1964;33:163–9.

quality improvement project. J Perinat Med 2013;41:455–60.

e198 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

83. Bartels DB, Kreienbrock L, Dammann O, Wenzlaff P, Poets CF. Population based 115. Petrova A, Demissie K, Rhoads GG, Smulian JC, Marcella S, Ananth CV. Associ-

study on the outcome of small for gestational age newborns. Arch Dis Child ation of maternal fever during labor with neonatal and infant morbidity and

Fetal Neonatal Ed 2005;90:F53–9. mortality. Obstet Gynecol 2001;98:20–7.

84. Carroll PD, Nankervis CA, Giannone PJ, Cordero L. Use of polyethylene bags in 116. Alexander JM, McIntire DM, Leveno KJ. Chorioamnionitis and the prognosis for

extremely low birth weight infant resuscitation for the prevention of hypother- term infants. Obstet Gynecol 1999;94:274–8.

mia. J Reprod Med 2010;55:9–13. 117. Greenwell EA, Wyshak G, Ringer SA, Johnson LC, Rivkin MJ, Lieberman E. Intra-

85. Gleissner M, Jorch G, Avenarius S. Risk factors for intraventricular hemor- partum temperature elevation, epidural use, and adverse outcome in term

rhage in a birth cohort of 3721 premature infants. J Perinat Med 2000;28: infants. Pediatrics 2012;129:e447–54.

104–10. 118. Goetzl L, Manevich Y, Roedner C, Praktish A, Hebbar L, Townsend DM. Maternal

86. Herting E, Speer CP, Harms K, et al. Factors influencing morbidity and mortality and fetal oxidative stress and intrapartum term fever. Am J Obstet Gynecol

in infants with severe respiratory distress syndrome treated with single or 2010;202:363.e1–5.

multiple doses of a natural porcine surfactant. Biol Neonate 1992;61:26–30. 119. Glass HC, Pham TN, Danielsen B, Towner D, Glidden D, Wu YW. Antenatal and

87. Van de Bor M, Van Bel F, Lineman R, Ruys JH. Perinatal factors and intrapartum risk factors for seizures in term newborns: a population-based

periventricular–intraventricular hemorrhage in preterm infants. Am J Dis Child study, California 1998–2002. J Pediatr 2009;154:24.e1–5.

1986;140:1125–30. 120. Lieberman E, Lang J, Richardson DK, Frigoletto FD, Heffner LJ, Cohen A. Intra-

88. Audeh S, Smolkin T, Bental Y, et al. Does admission hypothermia predispose to partum maternal fever and neonatal outcome. Pediatrics 2000;105:8–13.

intraventricular hemorrhage in very-low-birth-weight infants? Neonatology 121. Lieberman E, Eichenwald E, Mathur G, Richardson D, Heffner L, Cohen

2011;100:373–9. A. Intrapartum fever and unexplained seizures in term infants. Pediatrics

89. Dincsoy MY, Siddiq F, Kim YM. Intracranial hemorrhage in hypothermic low- 2000;106:983–8.

birth-weight neonates. Childs Nerv Syst 1990;6:245–8 [discussion 248]. 122. Badawi N, Kurinczuk JJ, Keogh JM, et al. Intrapartum risk factors for

90. Levene MI, Fawer CL, Lamont RF. Risk factors in the development of intraven- newborn encephalopathy: the Western Australian case–control study. BMJ

tricular haemorrhage in the preterm neonate. Arch Dis Child 1982;57:410–7. 1998;317:1554–8.

91. Rong Z, Liu H, Xia S, Chang L. Risk and protective factors of intraventricular hem- 123. Impey L, Greenwood C, MacQuillan K, Reynolds M, Sheil O. Fever in labour

orrhage in preterm babies in Wuhan, China. Childs Nerv Syst 2012;28:2077–84. and neonatal encephalopathy: a prospective cohort study. BJOG 2001;108:

92. Szymonowicz W, Yu VY, Wilson FE. Antecedents of periventricular haemorr- 594–7.

hage in infants weighing 1250 g or less at birth. Arch Dis Child 1984;59:13–7. 124. Impey LW, Greenwood CE, Black RS, Yeh PS, Sheil O, Doyle P. The relationship

93. DeMauro SB, Douglas E, Karp K, et al. Improving delivery room management between intrapartum maternal fever and neonatal acidosis as risk factors for

for very preterm infants. Pediatrics 2013;132:e1018–25. neonatal encephalopathy. Am J Obstet Gynecol 2008;198:49.e1–6.

94. Harms K, Herting E, Kron M, Schill M, Schiffmann H. Importance of pre- 125. Linder N, Fridman E, Makhoul A, et al. Management of term newborns fol-

and perinatal risk factors in respiratory distress syndrome of premature lowing maternal intrapartum fever. J Matern Fetal Neonatal Med 2013;26:

infants. A logical regression analysis of 1100 cases. Z Geburtshilfe Neonatol 207–10.

1997;201:258–62. 126. Butwick AJ, Lipman SS, Carvalho B. Intraoperative forced air-warming during

95. Lee HC, Powers RJ, Bennett MV, et al. Implementation methods for deliv- cesarean delivery under spinal anesthesia does not prevent maternal hypother-

ery room management: a quality improvement comparison study. Pediatrics mia. Anesth Analg 2007;105:1413–9 [table of contents].

2014;134:e1378–86. 127. Fallis WM, Hamelin K, Symonds J, Wang X. Maternal and newborn outcomes

96. Russo A, McCready M, Torres L, et al. Reducing hypothermia in preterm infants related to maternal warming during cesarean delivery. J Obstet Gynecol Neona-

following delivery. Pediatrics 2014;133:e1055–62. tal Nurs 2006;35:324–31.

97. Anderson S, Shakya KN, Shrestha LN, Costello AM. Hypoglycaemia: a com- 128. Horn EP, Schroeder F, Gottschalk A, et al. Active warming during cesarean

mon problem among uncomplicated newborn infants in Nepal. J Trop Pediatr delivery. Anesth Analg 2002;94:409–14 [table of contents].

1993;39:273–7. 129. Woolnough M, Allam J, Hemingway C, Cox M, Yentis SM. Intra-operative fluid

98. Lazic-Mitrovi´ c´ T, Djukic´ M, Cutura N, et al. Transitory hypothermia as early warming in elective caesarean section: a blinded randomised controlled trial.

prognostic factor in term newborns with intrauterine growth retardation. Srp Int J Obstet Anesth 2009;18:346–51.

Arh Celok Lek 2010;138:604–8. 130. Yokoyama K, Suzuki M, Shimada Y, Matsushima T, Bito H, Sakamoto A. Effect

99. Lenclen R, Mazraani M, Jugie M, et al. Use of a polyethylene bag: a way to of administration of pre-warmed intravenous fluids on the frequency of

improve the thermal environment of the premature newborn at the delivery hypothermia following spinal anesthesia for Cesarean delivery. J Clin Anesth

room. Arch Pediatr 2002;9:238–44. 2009;21:242–8.

100. Sasidharan CK, Gokul E, Sabitha S. Incidence and risk factors for neonatal hypo- 131. Belsches TC, Tilly AE, Miller TR, et al. Randomized trial of plastic bags to

glycaemia in Kerala, India. Ceylon Med J 2004;49:110–3. prevent term neonatal hypothermia in a resource-poor setting. Pediatrics

101. Mullany LC. Neonatal hypothermia in low-resource settings. Semin Perinatol 2013;132:e656–61.

2010;34:426–33. 132. Leadford AE, Warren JB, Manasyan A, et al. Plastic bags for preven-

102. McCarthy LK, Molloy EJ, Twomey AR, Murphy JF, O’Donnell CP. A randomized tion of hypothermia in preterm and low birth weight infants. Pediatrics

trial of exothermic mattresses for preterm newborns in polyethylene bags. 2013;132:e128–34.

Pediatrics 2013;132:e135–41. 133. Raman S, Shahla A. Temperature drop in normal term newborn infants

103. Chawla S, Amaram A, Gopal SP, Natarajan G. Safety and efficacy of Trans- born at the University Hospital, Kuala Lumpur. Aust N Z J Obstet Gynaecol

warmer mattress for preterm neonates: results of a randomized controlled 1992;32:117–9.

trial. J Perinatol 2011;31:780–4. 134. Fardig JA. A comparison of skin-to-skin contact and radiant heaters in promot-

104. Ibrahim CP, Yoxall CW. Use of self-heating gel mattresses eliminates admis- ing neonatal thermoregulation. J Nurse Midwifery 1980;25:19–28.

sion hypothermia in infants born below 28 weeks gestation. Eur J Pediatr 135. Christensson K, Siles C, Moreno L, et al. Temperature, metabolic adaptation

2010;169:795–9. and crying in healthy full-term newborns cared for skin-to-skin or in a cot.

105. Singh A, Duckett J, Newton T, Watkinson M. Improving neonatal unit admission Acta Paediatr 1992;81:488–93.

temperatures in preterm babies: exothermic mattresses, polythene bags or a 136. Christensson K. Fathers can effectively achieve heat conservation in healthy

traditional approach? J Perinatol 2010;30:45–9. newborn infants. Acta Paediatr 1996;85:1354–60.

106. McCarthy LK, O’Donnell CP. Warming preterm infants in the delivery 137. Bystrova K, Widström AM, Matthiesen AS, et al. Skin-to-skin contact may

room: polyethylene bags, exothermic mattresses or both? Acta Paediatr reduce negative consequences of “the stress of being born”: a study on temper-

2011;100:1534–7. ature in newborn infants, subjected to different ward routines in St. Petersburg.

107. Knobel RB, Wimmer Jr JE, Holbert D. Heat loss prevention for preterm infants Acta Paediatr 2003;92:320–6.

in the delivery room. J Perinatol 2005;25:304–8. 138. Gouchon S, Gregori D, Picotto A, Patrucco G, Nangeroni M, Di Giulio P. Skin-

108. te Pas AB, Lopriore E, Dito I, Morley CJ, Walther FJ. Humidified and heated air to-skin contact after cesarean delivery: an experimental study. Nurs Res

during stabilization at birth improves temperature in preterm infants. Pedi- 2010;59:78–84.

atrics 2010;125:e1427–32. 139. Marín Gabriel MA, Llana Martín I, López Escobar A, Fernández Villalba E,

109. Doglioni N, Cavallin F, Mardegan V, et al. Total body polyethylene wraps Romero Blanco I, Touza Pol P. Randomized controlled trial of early skin-

for preventing hypothermia in preterm infants: a randomized trial. J Pediatr to-skin contact: effects on the mother and the newborn. Acta Paediatr

2014;165:261.e1–6. 2010;99:1630–4.

110. Pinheiro JM, Furdon SA, Boynton S, Dugan R, Reu-Donlon C, Jensen S. Decreasing 140. Nimbalkar SM, Patel VK, Patel DV, Nimbalkar AS, Sethi A, Phatak A. Effect of

hypothermia during delivery room stabilization of preterm neonates. Pedi- early skin-to-skin contact following normal delivery on incidence of hypother-

atrics 2014;133:e218–26. mia in neonates more than 1800g: randomized control trial. J Perinatol

111. Motil KJ, Blackburn MG, Pleasure JR. The effects of four different radi- 2014;34:364–8.

ant warmer temperature set-points used for rewarming neonates. J Pediatr 141. Bergman NJ, Linley LL, Fawcus SR. Randomized controlled trial of skin-to-skin

1974;85:546–50. contact from birth versus conventional incubator for physiological stabilization

112. Tafari N, Gentz J. Aspects of rewarming newborn infants with severe accidental in 1200- to 2199-gram newborns. Acta Paediatr 2004;93:779–85.

hypothermia. Acta Paediatr Scand 1974;63:595–600. 142. Morley CJ, Davis PG, Doyle LW, et al. Nasal CPAP or intubation at birth for very

113. Racine J, Jarjoui E. Severe hypothermia in infants. Helv Paediatr Acta preterm infants. N Engl J Med 2008;358:700–8.

1982;37:317–22. 143. Support Study Group of the Eunice Kennedy Shriver NICHD Neonatal Research

114. Sofer S, Yagupsky P, Hershkowits J, Bearman JE. Improved outcome of Network, Finer NN, Carlo WA, et al. Early CPAP versus surfactant in extremely

hypothermic infants. Pediatr Emerg Care 1986;2:211–4. preterm infants. N Engl J Med 2010;362:1970–9.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e199

144. Dunn MS, Kaempf J, de Klerk A, et al. Randomized trial comparing 3 174. Armanian AM, Badiee Z. Resuscitation of preterm newborns with low con-

approaches to the initial respiratory management of preterm neonates. Pedi- centration oxygen versus high concentration oxygen. J Res Pharm Pract

atrics 2011;128:e1069–76. 2012;1:25–9.

145. Dani C, Bertini G, Pezzati M, Cecchi A, Caviglioli C, Rubaltelli FF. Early extubation 175. Kapadia VS, Chalak LF, Sparks JE, Allen JR, Savani RC, Wyckoff MH. Resuscita-

and nasal continuous positive airway pressure after surfactant treatment for tion of preterm neonates with limited versus high oxygen strategy. Pediatrics

respiratory distress syndrome among preterm infants <30 weeks’ gestation. 2013;132:e1488–96.

Pediatrics 2004;113:e560–3. 176. Lundstrøm KE, Pryds O, Greisen G. Oxygen at birth and prolonged cere-

146. Vyas H, Milner AD, Hopkin IE, Boon AW. Physiologic responses to prolonged bral vasoconstriction in preterm infants. Arch Dis Child Fetal Neonatal Ed

and slow-rise inflation in the resuscitation of the asphyxiated newborn infant. 1995;73:F81–6.

J Pediatr 1981;99:635–9. 177. Rabi Y, Singhal N, Nettel-Aguirre A. Room-air versus oxygen adminis-

147. Harling AE, Beresford MW, Vince GS, Bates M, Yoxall CW. Does sustained lung tration for resuscitation of preterm infants: the ROAR study. Pediatrics

inflation at resuscitation reduce lung injury in the preterm infant? Arch Dis 2011;128:e374–81.

Child Fetal Neonatal Ed 2005;90:F406–10. 178. Rook D, Schierbeek H, Vento M, et al. Resuscitation of preterm infants with

148. Lindner W, Högel J, Pohlandt F. Sustained pressure-controlled inflation or inter- different inspired oxygen fractions. J Pediatr 2014;164:1322-6.e3.

mittent mandatory ventilation in preterm infants in the delivery room? A 179. Vento M, Moro M, Escrig R, et al. Preterm resuscitation with low oxygen

randomized, controlled trial on initial respiratory support via nasopharyngeal causes less oxidative stress, inflammation, and chronic lung disease. Pediatrics

tube. Acta Paediatr 2005;94:303–9. 2009;124:e439–49.

149. Lista G, Fontana P, Castoldi F, Cavigioli F, Dani C. Does sustained lung inflation 180. Wang CL, Anderson C, Leone TA, Rich W, Govindaswami B, Finer NN. Resus-

at birth improve outcome of preterm infants at risk for respiratory distress citation of preterm neonates by using room air or 100% oxygen. Pediatrics

syndrome? Neonatology 2011;99:45–50. 2008;121:1083–9.

150. Lindner W, Vossbeck S, Hummler H, Pohlandt F. Delivery room management 181. Dawson JA, Kamlin CO, Wong C, et al. Oxygen saturation and heart

of extremely low birth weight infants: spontaneous breathing or intubation? rate during delivery room resuscitation of infants <30 weeks’ gestation

Pediatrics 1999;103:961–7. with air or 100% oxygen. Arch Dis Child Fetal Neonatal Ed 2009;94:

151. Klingenberg C, Sobotka KS, Ong T, et al. Effect of sustained inflation duration; F87–91.

resuscitation of near-term asphyxiated lambs. Arch Dis Child Fetal Neonatal 182. Christman C, Hemway RJ, Wyckoff MH, Perlman JM. The two-thumb is supe-

Ed 2013;98:F222–7. rior to the two-finger method for administering chest compressions in a

152. te Pas AB, Siew M, Wallace MJ, et al. Effect of sustained inflation length on estab- manikin model of neonatal resuscitation. Arch Dis Child Fetal Neonatal Ed

lishing functional residual capacity at birth in ventilated premature rabbits. 2011;96:F99–101.

Pediatr Res 2009;66:295–300. 183. Dorfsman ML, Menegazzi JJ, Wadas RJ, Auble TE. Two-thumb vs. two-finger

153. Klöpping-Ketelaars WA, Maertzdorf WJ, Blanco CE. Effect of sustained infla- chest compression in an infant model of prolonged cardiopulmonary resusci-

tions applied directly after cord clamping on lung function in premature tation. Acad Emerg Med 2000;7:1077–82.

newborn lambs. Acta Paediatr 1994;83:1017–21. 184. Houri PK, Frank LR, Menegazzi JJ, Taylor R. A randomized, controlled trial of

154. Tingay DG, Bhatia R, Schmölzer GM, Wallace MJ, Zahra VA, Davis PG. Effect two-thumb vs two-finger chest compression in a swine infant model of cardiac

of sustained inflation vs. stepwise PEEP strategy at birth on gas exchange and arrest [see comment]. Prehosp Emerg Care 1997;1:65–7.

lung mechanics in preterm lambs. Pediatr Res 2014;75:288–94. 185. Martin PS, Kemp AM, Theobald PS, Maguire SA, Jones MD. Do chest

155. te Pas AB, Walther FJ. A randomized, controlled trial of delivery room compressions during simulated infant CPR comply with international recom-

respiratory management in very preterm infants. Pediatrics 2007;120: mendations? Arch Dis Child 2013;98:576–81.

322–9. 186. Martin PS, Kemp AM, Theobald PS, Maguire SA, Jones MD. Does a more “phys-

156. Dawson JA, Schmölzer GM, Kamlin CO, et al. Oxygenation with T-piece versus iological” infant manikin design effect chest compression quality and create a

self-inflating bag for ventilation of extremely preterm infants at birth: a ran- potential for thoracic over-compression during simulated infant CPR? Resus-

domized controlled trial. J Pediatr 2011;158:912.e1–7. citation 2013;84:666–71.

157. Szyld E, Aguilar A, Musante GA, et al. Comparison of devices for newborn 187. Martin P, Theobald P, Kemp A, Maguire S, Maconochie I, Jones M. Real-

ventilation in the delivery room. J Pediatr 2014;165:234.e3–8. time feedback can improve infant manikin cardiopulmonary resuscitation

158. Al Takroni AM, Parvathi CK, Mendis KB, Hassan S, Reddy I, Kudair HA. Selective by up to 79% – a randomised controlled trial. Resuscitation 2013;84:

tracheal suctioning to prevent meconium aspiration syndrome. Int J Gynaecol 1125–30.

Obstet 1998;63:259–63. 188. Menegazzi JJ, Auble TE, Nicklas KA, Hosack GM, Rack L, Goode JS. Two-thumb

159. Peng TC, Gutcher GR, Van Dorsten JP. A selective aggressive approach to the versus two-finger chest compression during CRP in a swine infant model of

neonate exposed to meconium-stained amniotic fluid. Am J Obstet Gynecol cardiac arrest. Ann Emerg Med 1993;22:240–3.

1996;175:296–301 [discussion 301]. 189. Udassi S, Udassi JP, Lamb MA, et al. Two-thumb technique is superior to

160. Yoder BA. Meconium-stained amniotic fluid and respiratory complications: two-finger technique during lone rescuer infant manikin CPR. Resuscitation

impact of selective tracheal suction. Obstet Gynecol 1994;83:77–84. 2010;81:712–7.

161. Chishty AL, Alvi Y, Iftikhar M, Bhutta TI. Meconium aspiration in neonates: 190. Whitelaw CC, Slywka B, Goldsmith LJ. Comparison of a two-finger versus two-

combined obstetric and paediatric intervention improves outcome. J Pak Med thumb method for chest compressions by healthcare providers in an infant

Assoc 1996;46:104–8. mechanical model. Resuscitation 2000;43:213–6.

162. Falciglia HS. Failure to prevent meconium aspiration syndrome. Obstet Gynecol 191. David R. Closed chest cardiac massage in the newborn infant. Pediatrics

1988;71:349–53. 1988;81:552–4.

163. Fox WW, Gutsche BB, DeVore JS. A delivery room approach to the meconium 192. Moya F, James LS, Burnard ED, Hanks EC. Cardiac massage in the newborn infant

aspiration syndrome (MAS). Immediate intubation, endotracheal suction, and through the intact chest. Am J Obstet Gynecol 1962;84:798–803.

oxygen administration can reduce morbidity and mortality. Clin Pediatr (Phila) 193. Thaler MM, Stobie GH. An improved technic of external cardiac compression

1977;16:325–8. in infants and young children. N Engl J Med 1963;269:606–10.

164. Gregory GA, Gooding CA, Phibbs RH, Tooley WH. Meconium aspiration in 194. Todres ID, Rogers MC. Methods of external cardiac massage in the newborn

infants – a prospective study. J Pediatr 1974;85:848–52. infant. J Pediatr 1975;86:781–2.

165. Wiswell TE, Tuggle JM, Turner BS. Meconium aspiration syndrome: have we 195. Dellimore K, Heunis S, Gohier F, et al. Development of a diagnostic glove for

made a difference? Pediatrics 1990;85:715–21. unobtrusive measurement of chest compression force and depth during neona-

166. Wiswell TE, Henley MA. Intratracheal suctioning, systemic infection, and the tal CPR. Conf Proc IEEE Eng Med Biol Soc 2013;2013:350–3.

meconium aspiration syndrome. Pediatrics 1992;89:203–6. 196. Park J, Yoon C, Lee JC, et al. Manikin-integrated digital measuring system for

167. Ting P, Brady JP. Tracheal suction in meconium aspiration. Am J Obstet Gynecol assessment of infant cardiopulmonary resuscitation techniques. IEEE J Biomed

1975;122:767–71. Health Inform 2014;18:1659–67.

168. Davis RO, Philips 3rd JB, Harris Jr BA, Wilson ER, Huddleston JF. Fatal meco- 197. Huynh TK, Hemway RJ, Perlman JM. The two-thumb technique using an ele-

nium aspiration syndrome occurring despite airway management considered vated surface is preferable for teaching infant cardiopulmonary resuscitation.

appropriate. Am J Obstet Gynecol 1985;151:731–6. J Pediatr 2012;161:658–61.

169. Dooley SL, Pesavento DJ, Depp R, Socol ML, Tamura RK, Wiringa KS. Meconium 198. Udassi JP, Udassi S, Theriaque DW, Shuster JJ, Zaritsky AL, Haque IU. Effect

below the vocal cords at delivery: correlation with intrapartum events. Am J of alternative chest compression techniques in infant and child on rescuer

Obstet Gynecol 1985;153:767–70. performance. Pediatr Crit Care Med 2009;10:328–33.

170. Hageman JR, Conley M, Francis K, et al. Delivery room management of meco- 199. Fakhraddin BZ, Shimizu N, Kurosawa S, Sakai H, Miyasaka K, Mizutani S. New

nium staining of the amniotic fluid and the development of meconium method of chest compression for infants in a single rescuer situation: thumb-

aspiration syndrome. J Perinatol 1988;8:127–31. index finger technique. J Med Dent Sci 2011;58:15–22.

171. Manganaro R, Mamì C, Palmara A, Paolata A, Gemelli M. Incidence of meconium 200. Udassi JP, Udassi S, Lamb MA, et al. Improved chest recoil using an adhesive

aspiration syndrome in term meconium-stained babies managed at birth with glove device for active compression-decompression CPR in a pediatric manikin

selective tracheal intubation. J Perinat Med 2001;29:465–8. model. Resuscitation 2009;80:1158–63.

172. Rossi EM, Philipson EH, Williams TG, Kalhan SC. Meconium aspiration 201. Franke I, Pingen A, Schiffmann H, et al. Cardiopulmonary resuscitation (CPR)-

syndrome: intrapartum and neonatal attributes. Am J Obstet Gynecol related posterior rib fractures in neonates and infants following recommended

1989;161:1106–10. changes in CPR techniques. Child Abuse Negl 2014;38:1267–74.

173. Suresh GK, Sarkar S. Delivery room management of infants born through thin 202. Orlowski JP. Optimum position for external cardiac compression in infants and

meconium stained liquor. Indian Pediatr 1994;31:1177–81. young children. Ann Emerg Med 1986;15:667–73.

e200 J. Wyllie et al. / Resuscitation 95 (2015) e169–e201

203. You Y. Optimum location for chest compressions during two-rescuer infant 231. Esmail N, Saleh M, Ali A. Laryngeal mask airway versus endotracheal intubation

cardiopulmonary resuscitation. Resuscitation 2009;80:1378–81. for Apgar score improvement in neonatal resuscitation. Egypt J Anesthesiol

204. Saini SS, Gupta N, Kumar P, Bhalla AK, Kaur H. A comparison of two-fingers 2002;18:115–21.

technique and two-thumbs encircling hands technique of chest compression 232. Schmölzer GM, Morley CJ, Wong C, et al. Respiratory function monitor guid-

in neonates. J Perinatol 2012;32:690–4. ance of mask ventilation in the delivery room: a feasibility study. J Pediatr

205. Phillips GW, Zideman DA. Relation of infant heart to sternum: its significance 2012;160:377-381.e2.

in cardiopulmonary resuscitation. Lancet 1986;1:1024–5. 233. Kong JY, Rich W, Finer NN, Leone TA. Quantitative end-tidal carbon diox-

206. Lee SH, Cho YC, Ryu S, et al. A comparison of the area of chest compression ide monitoring in the delivery room: a randomized controlled trial. J Pediatr

by the superimposed-thumb and the alongside-thumb techniques for infant 2013;163:104-8.e1.

cardiopulmonary resuscitation. Resuscitation 2011;82:1214–7. 234. Berg RA, Henry C, Otto CW, et al. Initial end-tidal CO2 is markedly elevated

207. Lim JS, Cho Y, Ryu S, et al. Comparison of overlapping (OP) and adjacent thumb during cardiopulmonary resuscitation after asphyxial cardiac arrest. Pediatr

positions (AP) for cardiac compressions using the encircling method in infants. Emerg Care 1996;12:245–8.

Emerg Med J 2013;30:139–42. 235. Chalak LF, Barber CA, Hynan L, Garcia D, Christie L, Wyckoff MH. End-tidal

208. Lee KH, Kim EY, Park DH, et al. Evaluation of the 2010 American Heart Associ- CO2 detection of an audible heart rate during neonatal cardiopulmonary

ation Guidelines for infant CPR finger/thumb positions for chest compression: resuscitation after asystole in asphyxiated piglets. Pediatr Res 2011;69:

a study using computed tomography. Resuscitation 2013;84:766–9. 401–5.

209. Clements F, McGowan J. Finger position for chest compressions in cardiac arrest 236. Bhende MS, Karasic DG, Menegazzi JJ. Evaluation of an end-tidal CO2 detector

in infants. Resuscitation 2000;44:43–6. during cardiopulmonary resuscitation in a canine model for pediatric cardiac

210. Solevåg AL, Dannevig I, Wyckoff M, Saugstad OD, Nakstad B. Extended series arrest. Pediatr Emerg Care 1995;11:365–8.

of cardiac compressions during CPR in a swine model of perinatal asphyxia. 237. Bhende MS, Karasic DG, Karasic RB. End-tidal carbon dioxide changes during

Resuscitation 2010;81:1571–6. cardiopulmonary resuscitation after experimental asphyxial cardiac arrest. Am

211. Solevåg AL, Dannevig I, Wyckoff M, Saugstad OD, Nakstad B. Return of spon- J Emerg Med 1996;14:349–50.

taneous circulation with a compression:ventilation ratio of 15:2 versus 3:1 in 238. Bhende MS, Thompson AE. Evaluation of an end-tidal CO2 detector during

newborn pigs with cardiac arrest due to asphyxia. Arch Dis Child Fetal Neonatal pediatric cardiopulmonary resuscitation. Pediatrics 1995;95:395–9.

Ed 2011;96:F417–21. 239. Jacobs SE, Morley CJ, Inder TE, et al. Whole-body hypothermia for term and

212. Dannevig I, Solevåg AL, Saugstad OD, Nakstad B. Lung injury in asphyxiated near-term newborns with hypoxic–ischemic encephalopathy: a randomized

newborn pigs resuscitated from cardiac arrest – the impact of supplemen- controlled trial. Arch Pediatr Adolesc Med 2011;165:692–700.

tary oxygen, longer ventilation intervals and chest compressions at different 240. Bharadwaj SK, Bhat BV. Therapeutic hypothermia using gel packs for term

compression-to-ventilation ratios. Open Respir Med J 2012;6:89–96. neonates with hypoxic ischaemic encephalopathy in resource-limited settings:

213. Dannevig I, Solevåg AL, Sonerud T, Saugstad OD, Nakstad B. Brain inflammation a randomized controlled trial. J Trop Pediatr 2012;58:382–8.

induced by severe asphyxia in newborn pigs and the impact of alternative 241. Robertson NJ, Hagmann CF, Acolet D, et al. Pilot randomized trial of therapeutic

resuscitation strategies on the newborn central nervous system. Pediatr Res hypothermia with serial cranial ultrasound and 18–22 month follow-up for

2013;73:163–70. neonatal encephalopathy in a low resource hospital setting in Uganda: study

214. Hemway RJ, Christman C, Perlman J. The 3:1 is superior to a 15:2 ratio in a protocol. Trials 2011;12:138.

newborn manikin model in terms of quality of chest compressions and number 242. Thayyil S, Shankaran S, Wade A, et al. Whole-body cooling in neonatal ence-

of ventilations. Arch Dis Child Fetal Neonatal Ed 2013;98:F42–5. phalopathy using phase changing material. Arch Dis Child Fetal Neonatal Ed

215. Solevåg AL, Madland JM, Gjærum E, Nakstad B. Minute ventilation at different 2013;98:F280–1.

compression to ventilation ratios, different ventilation rates, and continuous 243. Medlock S, Ravelli AC, Tamminga P, Mol BW, Abu-Hanna A. Prediction of mor-

chest compressions with asynchronous ventilation in a newborn manikin. tality in very premature infants: a systematic review of prediction models. PLoS

Scand J Trauma Resusc Emerg Med 2012;20:73. ONE 2011;6:e23441.

216. Lakshminrusimha S, Steinhorn RH, Wedgwood S, et al. Pulmonary hemody- 244. Tyson JE, Parikh NA, Langer J, Green C, Higgins RD, National Institute of

namics and vascular reactivity in asphyxiated term lambs resuscitated with 21 Child Health and Human Development Neonatal Research Network. Intensive

and 100% oxygen. J Appl Physiol (1985) 2011;111:1441–7. care for extreme prematurity-moving beyond gestational age. N Engl J Med

217. Linner R, Werner O, Perez-de-Sa V, Cunha-Goncalves D. Circulatory recovery is 2008;358:1672–81.

as fast with air ventilation as with 100% oxygen after asphyxia-induced cardiac 245. Manktelow BN, Seaton SE, Field DJ, Draper ES. Population-based estimates of

arrest in piglets. Pediatr Res 2009;66:391–4. in-unit survival for very preterm infants. Pediatrics 2013;131:e425–32.

218. Lipinski CA, Hicks SD, Callaway CW. Normoxic ventilation during resus- 246. Ambalavanan N, Carlo WA, Bobashev G, et al. Prediction of death for extremely

citation and outcome from asphyxial cardiac arrest in rats. Resuscitation low birth weight neonates. Pediatrics 2005;116:1367–73.

1999;42:221–9. 247. Bottoms SF, Paul RH, Mercer BM, et al. Obstetric determinants of neonatal sur-

219. Perez-de-Sa V, Cunha-Goncalves D, Nordh A, et al. High brain tissue oxygen vival: antenatal predictors of neonatal survival and morbidity in extremely low

tension during ventilation with 100% oxygen after fetal asphyxia in newborn birth weight infants. Am J Obstet Gynecol 1999;180:665–9.

sheep. Pediatr Res 2009;65:57–61. 248. Casalaz DM, Marlow N, Speidel BD. Outcome of resuscitation following

220. Solevåg AL, Dannevig I, Nakstad B, Saugstad OD. Resuscitation of severely unexpected apparent stillbirth. Arch Dis Child Fetal Neonatal Ed 1998;78:

asphyctic newborn pigs with cardiac arrest by using 21% or 100% oxygen. F112–5.

Neonatology 2010;98:64–72. 249. Harrington DJ, Redman CW, Moulden M, Greenwood CE. The long-term

221. Temesvári P, Karg E, Bódi I, et al. Impaired early neurologic outcome in new- outcome in surviving infants with Apgar zero at 10 minutes: a systematic

born piglets reoxygenated with 100% oxygen compared with room air after review of the literature and hospital-based cohort. Am J Obstet Gynecol

pneumothorax-induced asphyxia. Pediatr Res 2001;49:812–9. 2007;196:463.e1–5.

222. Walson KH, Tang M, Glumac A, et al. Normoxic versus hyperoxic resuscitation 250. Kasdorf E, Laptook A, Azzopardi D, Jacobs S, Perlman JM. Improving infant

in pediatric asphyxial cardiac arrest: effects on oxidative stress. Crit Care Med outcome with a 10 min Apgar of 0. Arch Dis Child Fetal Neonatal Ed

2011;39:335–43. 2015;100:F102–5.

223. Yeh ST, Cawley RJ, Aune SE, Angelos MG. Oxygen requirement during car- 251. Laptook AR, Shankaran S, Ambalavanan N, et al. Outcome of term infants using

diopulmonary resuscitation (CPR) to effect return of spontaneous circulation. Apgar scores at 10 minutes following hypoxic–ischemic encephalopathy. Pedi-

Resuscitation 2009;80:951–5. atrics 2009;124:1619–26.

224. Matsiukevich D, Randis TM, Utkina-Sosunova I, Polin RA, Ten VS. The state of 252. Patel H, Beeby PJ. Resuscitation beyond 10 minutes of term babies born without

systemic circulation, collapsed or preserved defines the need for hyperoxic or signs of life. J Paediatr Child Health 2004;40:136–8.

normoxic resuscitation in neonatal mice with hypoxia–ischemia. Resuscitation 253. Sarkar S, Bhagat I, Dechert RE, Barks JD. Predicting death despite therapeu-

2010;81:224–9. tic hypothermia in infants with hypoxic–ischaemic encephalopathy. Arch Dis

225. Faa A, Iacovidou N, Xanthos T, et al. Hypoxia/reoxygenation-induced myocar- Child Fetal Neonatal Ed 2010;95:F423–8.

dial lesions in newborn piglets are related to interindividual variability and not 254. Nelson KB, Ellenberg JH. Apgar scores as predictors of chronic neurologic dis-

to oxygen concentration. Clinics (Sao Paulo) 2012;67:503–8. ability. Pediatrics 1981;68:36–44.

226. Markus T, Hansson S, Amer-Wåhlin I, Hellström-Westas L, Saugstad OD, Ley D. 255. Natarajan G, Shankaran S, Laptook AR, et al. Apgar scores at 10 min and out-

Cerebral inflammatory response after fetal asphyxia and hyperoxic resuscita- comes at 6–7 years following hypoxic–ischaemic encephalopathy. Arch Dis

tion in newborn sheep. Pediatr Res 2007;62:71–7. Child Fetal Neonatal Ed 2013;98:F473–9.

227. Mendoza-Paredes A, Liu H, Schears G, et al. Resuscitation with 100%, com- 256. Scott H. Outcome of very severe birth asphyxia. Arch Dis Child 1976;51:712–6.

pared with 21%, oxygen following brief, repeated periods of apnea can 257. Koppe JG, Kleiverda G. Severe asphyxia and outcome of survivors. Resuscitation

protect vulnerable neonatal brain regions from apoptotic injury. Resuscitation 1984;12:193–206.

2008;76:261–70. 258. Steiner H, Neligan G. Perinatal cardiac arrest. Quality of the survivors. Arch Dis

228. Singh R. Controlled trial to evaluate the use of LMA for neonatal resuscitation. Child 1975;50:696–702.

J Anaesthesiol Clin Pharmacol 2005;21:303–6. 259. Ekert P, Perlman M, Steinlin M, Hao Y. Predicting the outcome of postas-

229. Zhu XY, Lin BC, Zhang QS, Ye HM, Yu RJ. A prospective evaluation of the effi- phyxial hypoxic–ischemic encephalopathy within 4 hours of birth. J Pediatr

cacy of the laryngeal mask airway during neonatal resuscitation. Resuscitation 1997;131:613–7.

2011;82:1405–9. 260. Shah PS, Beyene J, To T, Ohlsson A, Perlman M. Postasphyxial hypoxic–ischemic

230. Feroze F, Masood N, Khuwaja A, Malik F. Neonatal resuscitation: the use of encephalopathy in neonates: outcome prediction rule within 4 hours of birth.

laryngeal mask airway. Prof Med J 2008;15:148–52. Arch Pediatr Adolesc Med 2006;160:729–36.

J. Wyllie et al. / Resuscitation 95 (2015) e169–e201 e201

261. Berden HJ, Willems FF, Hendrick JM, Pijls NH, Knape JT. How frequently should 272. Mosley CM, Shaw BN. A longitudinal cohort study to investigate the retention

basic cardiopulmonary resuscitation training be repeated to maintain adequate of knowledge and skills following attendance on the Newborn Life support

skills? BMJ 1993;306:1576–7. course. Arch Dis Child 2013;98:582–6.

262. Ernst KD, Cline WL, Dannaway DC, et al. Weekly and consecutive day neona- 273. Nadel FM, Lavelle JM, Fein JA, Giardino AP, Decker JM, Durbin DR. Teaching

tal intubation training: comparable on a pediatrics clerkship. Acad Med resuscitation to pediatric residents: the effects of an intervention. Arch Pediatr

2014;89:505–10. Adolesc Med 2000;154:1049–54.

263. Kaczorowski J, Levitt C, Hammond M, et al. Retention of neonatal resus- 274. Niles D, Sutton RM, Donoghue A, et al. “Rolling Refreshers”: a novel approach

citation skills and knowledge: a randomized controlled trial. Fam Med to maintain CPR psychomotor skill competence. Resuscitation 2009;80:

1998;30:705–11. 909–12.

264. Kovacs G, Bullock G, Ackroyd-Stolarz S, Cain E, Petrie D. A randomized 275. Nishisaki A, Donoghue AJ, Colborn S, et al. Effect of just-in-time simulation

controlled trial on the effect of educational interventions in promot- training on tracheal intubation procedure safety in the pediatric intensive care

ing airway management skill maintenance. Ann Emerg Med 2000;36: unit. Anesthesiology 2010;113:214–23.

301–9. 276. O’Donnell CM, Skinner AC. An evaluation of a short course in resuscitation

265. Montgomery C, Kardong-Edgren SE, Oermann MH, Odom-Maryon T. training in a district general hospital. Resuscitation 1993;26:193–201.

Student satisfaction and self report of CPR competency: HeartCode 277. Naji SA, Maguire GP, Fairbairn SA, Goldberg DP, Faragher EB. Training clinical

BLS courses, instructor-led CPR courses, and monthly voice advisory teachers in psychiatry to teach interviewing skills to medical students. Med

manikin practice for CPR skill maintenance. Int J Nurs Educ Scholarsh Educ 1986;20:140–7.

2012;9. 278. Breckwoldt J, Svensson J, Lingemann C, Gruber H. Does clinical teacher training

266. Oermann MH, Kardong-Edgren SE, Odom-Maryon T. Effects of monthly always improve teaching effectiveness as opposed to no teacher training? A

practice on nursing students’ CPR psychomotor skill performance. Resuscita- randomized controlled study. BMC Med Educ 2014;14:6.

tion 2011;82:447–53. 279. Boerboom TB, Jaarsma D, Dolmans DH, Scherpbier AJ, Mastenbroek NJ, Van

267. Stross JK. Maintaining competency in advanced cardiac life support skills. JAMA Beukelen P. Peer group reflection helps clinical teachers to critically reflect on

1983;249:3339–41. their teaching. Med Teach 2011;33:e615–23.

268. Su E, Schmidt TA, Mann NC, Zechnich AD. A randomized controlled trial to 280. Litzelman DK, Stratos GA, Marriott DJ, Lazaridis EN, Skeff KM. Beneficial and

assess decay in acquired knowledge among paramedics completing a pediatric harmful effects of augmented feedback on physicians’ clinical-teaching per-

resuscitation course. Acad Emerg Med 2000;7:779–86. formances. Acad Med 1998;73:324–32.

269. Sutton RM, Niles D, Meaney PA, et al. “Booster” training: evaluation 281. Schum TR, Yindra KJ. Relationship between systematic feedback to faculty and

of instructor-led bedside cardiopulmonary resuscitation skill training and ratings of clinical teaching. Acad Med 1996;71:1100–2.

automated corrective feedback to improve cardiopulmonary resuscitation 282. Skeff KM, Stratos G, Campbell M, Cooke M, Jones 3rd HW. Evaluation of the

compliance of Pediatric Basic Life Support providers during simulated cardiac seminar method to improve clinical teaching. J Gen Intern Med 1986;1:315–22.

arrest. Pediatr Crit Care Med 2011;12:e116–21. 283. Lye P, Heidenreich C, Wang-Cheng R, Bragg D, Simpson D, Advanced Fac-

270. Turner NM, Scheffer R, Custers E, Cate OT. Use of unannounced spaced tele- ulty Development Group. Experienced clinical educators improve their clinical

phone testing to improve retention of knowledge after life-support courses. teaching effectiveness. Ambul Pediatr 2003;3:93–7.

Med Teach 2011;33:731–7. 284. Regan-Smith M, Hirschmann K, Iobst W. Direct observation of faculty with

271. Lubin J, Carter R. The feasibility of daily mannequin practice to improve intu- feedback: an effective means of improving patient-centered and learner-

bation success. Air Med J 2009;28:195–7. centered teaching skills. Teach Learn Med 2007;19:278–86.